Professional Documents
Culture Documents
ISBN 0-9717288-0-1
October 2001
Published by the Department of Energy and the Petroleum Environmental Research
Forum.
Printed by the Government Printing Office
Additional copies of this report are available from:
Nancy Comstock
U.S. Department of Energy
National Energy Technology Laboratory
National Petroleum Technology Office
One West Third Street
Tulsa, OK 74103-3519
All rights reserved. No part of this book may be reproduced, stored in a retrieval system,
or transcribed in any form or by any means, electronic or mechanical, including
photocopying and recording, without the prior written permission of the publisher.
Printed in the United States of America
Table of Contents
iii
Preface
Preface
Remarkable changes in the approach used for protection of human health and
the environment in the United States have occurred over the past 30 years.
During the period from the early 1970s through the early 1990s, setting
priorities on which environmental problems were important was not an issue.
The problems and apparent culprits were obvious. The public and regulatory
agencies agreed on the severity of environmental problems and what was needed
to correct them. As a result, legislation was passed, regulations were
promulgated, and efforts and resources were focused on obvious pollution
control problems. Meeting these regulatory needs became the primary goal of
municipalities and industries during this period.
Over time, the increasing number of environmental laws and regulations, as well
as the increasing complexity of environmental problems, began to strain the
resources of regulatory agencies, municipalities, and industry. Interest increased
in identifying and considering different approaches for environmental
protection. As early as the mid-1980s, it was recognized that approaches other
than the compliance, command, and control approach for environmental
protection were needed. While he was Administrator of the United States
Environmental Protection Agency (USEPA) in the mid-1980s, William
Ruckelshaus attempted to introduce the concepts of risk assessment and risk
management into the USEPA decision-making process. Attempts also were
made to distinguish between the two concepts.
In 1990, then USEPA Administrator William Reilly called for a national debate
on national directions and policies for environmental protection. What
stimulated that call was the 1990 USEPA Science Advisory Board Report,
Reducing Risk: Setting Priorities and Strategies for Environmental Protection.
Among the many recommendations in that report was one that recommended
targeting environmental protection efforts on the basis of environmental risk and
risk reduction opportunities. Also recommended was the use of risk-based
priorities in environmental planning and budgeting.
Winston Churchill, at a critical turning point in World War II, stated, This is
not the end. It is not even the beginning of the end. But it is, perhaps the end of
the beginning. Unrecognized at the time, but obvious in hindsight, is the fact
that, in the United States, about the year 1990 was the end of the beginning
period of environmental protection in which government laws and regulations
were considered as the best approach for such protection. From the early
1990s, environmental protection efforts increasingly have focused on risk
assessment and risk management approaches.
Preface
It is clear that the compliance, command, and control approach that dominated
regulatory enforcement policy since the 1970s is no longer the predominant
driver of the profession and the environmental industry. It also is clear that we
will see a continuing firm regulatory baseline, with strong enforcement to define
the floor for environmental progress and sustain the legal mechanism for
penalizing environmental violators. Thus, we will have both a strong regulatory
program and an increasing emphasis on risk assessment and risk management
approaches for environmental protection.
From the initial efforts in the 1980s, risk-based evaluations and decisions have
become an important component of efforts to protect human health and the
environment. The information in Chapter 1 of this document provides an
excellent summary of the development of the risk-based decision-making
(RBDM) process and the current state of its use, particularly for sites with
contaminated soils. Thus, Chapter 1 is an overview of the important concepts
and incremental growth of RBDM in this second and important phase of
environmental protection.
A value of the RBDM process is that each site is treated individually and that
the remedial measures result in cleanup levels that are environmentally
acceptable for the given site characteristics and anticipated land use. For
situations where there are groups of sites, the RBDM process allows resources
to be focused on sites or areas that have greater environmental concerns. The
RBDM process also allows regulations to incorporate reasonable but
conservative Tier 1 screening levels. This avoids the use of overly
conservative approaches such as achieving background concentrations or
utilizing technologies deemed best available by non-risk based criteria.
Utilization of the RBDM process:
For sites with contaminated soils, the RBDM process results in achieving an
environmentally protective endpoint, i.e., a concentration of a chemical in such
soils below which there is no expected adverse effect to human health and the
environment.
The increased emphasis on RBDM has required greater knowledge of factors
that affect chemical release, transport, exposure, and general availability in the
environment. In turn, this need triggered scientific, pilot scale, and field
evaluations that have provided such knowledge.
Preface
Preface
Preface
analytical issues need to be discussed clearly and in a risk context. This book
provides such information in a logical and understandable manner. As such, it is
an important risk assessment and risk communication tool. It is my expectation
that future evaluations of E&P sites, and of other relevant sites, will be much
improved as a result of the information and logic presented in this book.
Raymond C. Loehr
H. M. Alharthy Centennial Chair
Environmental and Water
Resources Engineering Program
University of Texas at Austin
June 15, 2001
Acknowledgments
Acknowledgments
This book was prepared in conjunction with Petroleum Environmental Research
Forum (PERF) Project 97-08. The companies and trade associations that
supported this project include the American Petroleum Institute, Arthur D.
Little, the Canadian Association of Petroleum Producers, Chevron Research and
Technology Company, Equilon Enterprises LLC, ExxonMobil Corporation, Gas
Technology Institute, Texaco Incorporated, Unocal Corporation, and the United
States Department of Energy (DOE). Funding for this book was provided
through a grant from DOE/National Energy Technology Laboratory, under the
direction of Nancy Comstock of the National Petroleum Technology Office.
DOE provided research through its Lawrence Berkeley National Laboratory and
Pacific Northwest National Laboratory. Additional funding was provided to Dr.
William Rixey by the New Mexico Oil and Gas Association Total Petroleum
Hydrocarbon Working Group, chaired by Paul Peacock of Marathon Oil.
The authors and editors of this book include:
Acknowledgments
List of Abbreviations
List of Abbreviations
gw
unsat
as
org
ws
A
ADEC
ADL
Ag
API
Silver
American Petroleum Institute
As
ASTM
Arsenic
American Society for Testing and Materials
AT
Averaging time
Atm
Ba
Atmosphere
Barium
BCE
Be
BTEX
BW
C
Body weight
Chemical concentration
Cal-EPA
CCME
Cd
Cadmium
List of Abbreviations
CERCLA
CF
Co
COC
Cof
Co max
Cppmv
Cr
CR
Chromium
Contact rate
Cs
CSF
CSM
CT
CTPH
Cv
Cvf
Cvs
Cw
Cw max
Cu
CVAAS
D
DAFgw
DAFmix
DAFo
DAFsat
DAFunsat
Dair
Deff
Dwater
List of Abbreviations
DwT,S
DRO
E&P
EC
ED
Equivalent carbon
Exposure duration
EF
Exposure frequency
EPC
FAAS
FID
Foc
Flame-ionizing detector
Fraction of organic carbon in soil
Foil
FT
GC
GFAAS
GRO
GW
GW
GW Ing
Groundwater
Protective concentration limit for ingestion of groundwater
H
Hg
HI
HQ
Hazard Index
Hazard Quotient
HMPW
HSW
Infiltration rate
ICP-MS
ID
IRsoil
IRw
Partition coefficient
Kd
KD
KH
10
List of Abbreviations
Ko
Koc
Koil
Kow
Ks
KT,S
KT, unsat
LA
LCS
Louisiana
Laboratory control samples
LDF
LF
LFSW
Leaching factor
Leaching factor from soil to water
LOAEL
M
Lowest-observed-adverse-effect level
Soil-to-skin adherence factor
MADEP
MCL
MDL
MF
mg/kg
Modifying factor
Milligrams per kilogram
mg/L
MGP
Mo
Molybdenum
mol
MS
Mole
Mass spectrometry, or matrix spike
MSD
MSDS
M org
Ms
MT
Mv
Mw
11
List of Abbreviations
mV
Millivolts
MW
MW i
Molecular weight
Molecular weight of contaminant of interest
MW o
MW TPH
NA
NAPL
ND
Ni
NOAEL
Nickel
No-observed-adverse-effect level
NOW
O&G
ORNL
OSHA
P
PAH
Pb
Polyaromatic hydrocarbon
Lead
PCL
PERF
PF
PHC
ppb
Petroleum hydrocarbon
Part per billion
ppm
PQL
PRG
Q
QC
RAFd
RAFo
RBC
RBCA
12
List of Abbreviations
RBDM
RBSL
RBSLS-GW
RCRA
RES
RfC
Reference concentration
RfD
RfDo
Reference dose
Oral chronic reference dose
RIVM
RPD
RSD
RT
Solubility
SA
Sb
SCC
Se
Seff
Effective solubility
SFo
Si
SIM
SJV
Sn
Tin
SOC
SPE
SQG
SRM
SSL
SSTL
Temperature
THQ
TNRCC
TOC
13
List of Abbreviations
TPH
TPHCWG
TR
TRRP
TX
Infiltration rate
UCM
UF
Ugw
USEPA
USGS
UST
Vanadium
v
VF
Seepage velocity
Volatilization factor
VFp
VFs s
VP
Vapor pressure
W
X
Weight fraction
Mole fraction
Xmax
14
Executive Summary
Executive Summary
This book presents the scientific background material necessary to support a
risk-based decision-making (RBDM) approach for managing hydrocarbonimpacted soil at oil and natural gas industry exploration and production (E&P)
sites . Much of the information presented in the various chapters was generated
as part of a multi-year joint industry Petroleum Environmental Research Forum
(PERF) project that focused on improving the technical basis for environmental
management of E&P sites . The general RBDM approach is largely based on
guidance originally developed by the United States Environmental Protection
Agency (USEPA). That framework was later refined by the American Society
for Testing and Materials (ASTM) for evaluating individual chemicals and the
Total Petroleum Hydrocarbon Criteria Working Group (TPHCWG) and PERF
for complex petroleum hydrocarbon mixtures . The information and approach
described in this report have recently been used in the development of risk-based
regulatory progra ms in several oil and natural gas producing states in the United
States .
Detailed information about the composition of crude oils and gas condensates is
summarized and the total petroleum hydrocarbon (TPH) fractions and individual
chemicals of potential concern in these mixtures are identified. The fate,
transport, and toxicity information necessary for estimating potential risks to
human health is then described and used to generate risk-based screening levels
(RBSLs ) that can be used in a Tier 1, or screening level, risk evaluation. These
RBSLs represent soil concentrations that are protective of human health and are
calculated using exposure equations recognized by USEPA as providing
conservative estimates (i.e., lower than necessary for the protection of human
health) of acceptable concentrations in soil. The calculated Tier 1 RBSLs can
be compared to regulatory criteria to determine whether the criteria are
protective of human health, or in instances where there are no regulatory criteria,
the RBSLs can be used as limiting concentrations for TPH or specific chemicals
of concern in soil.
15
Executive Summary
exposure pathways of greatest concern are associated with direct human contact
with hydro carbon-impacted soil (i.e., soil ingestion, inhalation of soil particles,
and dermal contact). Leaching to groundwater and volatilization to outdoor air
are of lower concern for the complex mixtures as a whole.
The TPH RBSLs calculated for direct contact with soil impacted by this wide
variety of crude oils range from 35,000 to 67,000 mg/kg (3.5-6.7% by weight)
TPH for non-residential sites . TPH RBSLs for condensates range from 39,000
to 116,000 mg/kg (3.9-11.6% by weight) TPH. These values are significantly
greater than the TPH concentration of 10,000 mg/kg that has often been used as
the regulatory criterion for E&P sites . The TPH RBSLs for selected E&P
wastes are also determined for the same exposure scenario. These values were
very similar to those for the crude oils, ranging from 50,000 to 89,000 mg/kg
(5.0-8.9% by weight). Since these results represent RBSLs for all of the types
of oils produced around the world and because they have been shown to be
conservative for associated wastes and soil, they can be used to set TPH RBSLs
for crude oil and condensates present in soils at all E&P sites .
It is important to remember that these RBSLs apply to non-residential land use
conditions. If a particular site is expected to be used for residential purposes,
TPH RBSLs appropriate for that land use should be developed. Residential
RBSLs may be an order of magnitude or more lower than those appropriate for
non-residential land use.
16
Executive Summary
consideration at crude oil or condensate spill sites. Routine analysis for these
chemicals in soil at E&P sites is not necessary to ensure protection of human
health.
The understanding of the impact of benzene on the management of E&P sites is
continuing to evolve. Benzene is typically considered to be a chemical of
potential concern at gasoline release sites ; however, most upstream regulatory
programs do not routinely require benzene analysis of E&P site soils and do not
routinely set regulatory limits for benzene in soil. In this study, benzene was
detected in crude oils at a mean concentration of 1,340 mg/kg oil and in
condensates at a mean concentration of 10,300 mg/kg. However, some crude
oils contain little or no benzene. The potential for benzene to leach to
groundwater should be carefully evaluated at E&P sites where groundwater
protection is an important consideration and where the oil or condensate is
known to contain significant amounts of benzene. It may be appropriate to
directly analyze for benzene at the site (using USEPA 8260) and to derive
chemical-specific benzene RBSLs for groundwater protection. In deriving
appropriate benzene RBSLs, careful consideration should be given to relevant
environmental fate and transport processes, including volatilization and
biodegradation. It is now known that assumptions made in previous screening
level evaluations are overly conservative for benzene, because of its specific fate
and transport properties. Previously published benzene RBSLs using these
overly conservative assumptions may be 10 to 1000 times lower than necessary
to protect human health.
17
Executive Summary
18
Chapter 1
Introduction
Risk-based decision-making is the process of making environmental
management decisions based upon an assessment of the potential risks that
chemicals at a site may pose to human health and the environment. The United
States Environmental Protection Agency (USEPA) has developed a general
framework for risk-based decision making based upon human health
considerations, and has established general guidelines for determining what
constitutes acceptable risk to human health [1,2]. These guidelines can be used
to determine when some type of risk management action is required at a site.
Although the overall framework for risk-based decision-making was originally
developed for use at sites impacted by hazardous materials, in reality it is
equally applicable to all types of sites, including oil and gas industry exploration
and production (E&P) sites.
The general framework for risk-based decision-making was originally developed
by the USEPA largely in response to the requirements of the Comprehensive
Environmental Response and Contingency Liability Act of 1980 (CERCLA).
This framework has been refined over time and several tiered approaches to
risk-based decision-making have subsequently been developed. A major goal of
the framework is to make certain that management decisions for
environmentally impacted sites provide an adequate level of protection for
human health and the environment. Therefore, a health risk evaluation process
was developed and the overall risk characterization is used to guide site
management decisions.
Risk-Based Decision-Making at
Environmentally-Impacted Sites
Historically, regulatory programs in the United States have established
environmental management goals (i.e., cleanup levels) for chemicals of potential
concern at specific sites based on:
19
Chapter 1
However, since none of these goals is directly tied to the actual risks posed by
the chemicals of concern, there is no way to determine whether or not these
goals actually protect human health and the environment. In addition, there is
no way to determine the cost/benefit associated with achieving the management
goals listed above, since the benefit of the action cannot be determined. Without
any knowledge of the benefit resulting from a given action, there is no way to
prioritize actions to focus them on those problems where the greatest potential
for risk reduction exists. This could conceivably result in a portion of the public
being left at risk, and in the misallocation of both technical and financial
resources.
In contrast, risk-based approaches to site management clearly describe the
potential health benefits that might result from a particular environmental
management decision. Consequently, the actions that are taken at a site can be
evaluated and prioritized based on the actual reduction in risk that would be
achieved, and technical and financial resources can be allocated appropriately.
Like all technical methodologies and protocols, risk-based decision-making is
not necessarily applicable to every situation at every site. For example, there
may be instances where a risk-based assessment concludes that total petroleum
hydrocarbon (TPH) concentrations at a specific site do not pose a health risk.
However, these same concentrations may produce unsightly conditions that are
not aesthetically acceptable for the current and/or future land use. In these
instances, common sense should be used to guide site management decisions.
It is also important to think carefully about the assumptions that are made when
using risk-based decision-making for site management. Since there tends to be
limited data available to conduct a risk-based evaluation of a site, there is
generally a need to make some basic assumptions during the risk evaluation.
Examples of assumptions that need to be made may include the toxicity of the
materials in question or the duration and extent of potential exposures. In every
risk evaluation, it is important to understand the sensitivity of the risk-based
decisions to the assumptions made in order to determine how robust the
evaluation is and the circumstances that might justify the use of different
assumptions. The greatest criticism of risk-based decision-making in site
management is that it can be manipulated to produce any result that is desired by
the user. The primary defense to this criticism is to make certain that all
assumptions are technically justifiable, and to examine the sensitivity of the
outcome to the more critical of these assumptions.
20
Chapter 1
Hazard Identification
Exposure Assessment
Toxicity (or Dose-Response) Assessment
Risk Characterization
21
Chapter 1
Once the receptors have been identified, the next step is to determine how they
might be exposed. This is a process in which potentially complete exposure
pathways are identified. In identifying these complete exposure pathways, the
sources of the chemicals at the site are determined and the ways in which they
may move around in the environment and be transported to places at which
receptors might be exposed are considered. For example, if a crude oil is spilled
on soil at a site, a worker in the area may be exposed by direct skin contact with
the impacted soil. In addition, some of the components of the crude oil may
volatilize into air and be inhaled by the worker, or some components may
migrate through the soil into the groundwater, be transported to a drinking water
well at some distance from the site, and subsequently be ingested. The final step
of the exposure assessment is to quantify the potential exposure to receptors
using standardized intake equations.
Toxicity (Dose-Response) Assessment
The toxicity assessment answers the question What dose levels of the
chemicals of potential concern may produce adverse health effects in people or
other receptors? In the toxicity assessment, chemicals are usually evaluated
separately for their abilities to cause cancer and other adverse health effects. All
chemicals have the inherent ability to cause adverse health effects at some dose
level, but only certain chemicals have the ability to cause cancer. Most of the
available toxicological data for both carcinogenic and non-carcinogenic
chemicals have been generated in the laboratory using pure chemicals that have
been added to the food or water of rats or mice. One of the major challenges
associated with the use of these data is in extrapolating the results for individual
chemicals to situations in which mixtures of chemicals, such as crude oil, may
be of concern. A second challenge is in extrapolating the laboratory results
obtained in rodents treated with chemicals to situations in which people are
exposed to chemicals. In both cases, the USEPA includes uncertainty factors
into the analysis to make certain that the toxicity of the chemical or mixture of
chemicals is not underestimated.
Risk Characterization
The final step of the risk evaluation for a site is one in which the results of the
exposure assessment are combined with the toxicity assessment to quantify the
potential risks to human health and the environment. The result is a
conservative risk estimate that is likely to overestimate the true risks posed by
the site. In reality, the true risk will most likely be much lower than the
estimated risk.
22
Chapter 1
Hazard Identification
Exposure Assessment
Toxicity (or Dose-Response) Assessment
Risk Characterization
23
Chapter 1
ADL =
C x CR EF x ED
x
BW
AT
where:
ADL = Average dose level (mg/kg BW-day)
C = Chemical concentration (e.g., mg/kg-soil or mg/L-water)
CR = Contact rate or the amount of impacted medium contacted per event
(e.g., liters/day, kg/day)
EF = Exposure frequency (days/year)
ED = Exposure duration (years)
BW = Body weight of the receptor (kg)
AT = Averaging time of the exposure (days)
Derivation of Toxicological Dose-Response Factors
In estimating risk, the exposure estimate is combined with a toxicological doseresponse factor. The dose-response factor depends upon the chemical, the route
of exposure, and the health effect that is of concern (i.e., carcinogenic or noncarcinogenic). It is generally derived by USEPA [13], or other regulatory
agencies, and is made available to the public for use by risk assessors. The data
on which these factors are based are usually generated in laboratory studies
using animals. The dose-response factors derived from these data include
reference doses (RfDs) or inhalation reference concentrations (RfCs) for
evaluating non-carcinogenic effects and cancer slope factors for evaluating
carcinogenic effects as described below:
Calculation of Risk
The risk calculations for non-carcinogenic health effects are expressed in terms
of a unitless hazard quotient that is calculated using the following equation:
24
Chapter 1
Hazard Quotient=
The threshold level of acceptability for the Hazard Quotient that has been
established by the USEPA [1] is the value 1.0, although some state regulatory
agencies within the United States have established different levels of
acceptability. Hazard quotients greater than 1.0 typically require further site
analysis or some sort of site action.
The risk calculation for carcinogenic health effects is based on a somewhat
similar equation:
Risk = Average Lifetime Daily Dose (mg/kg-day) x Slope Factor (mg/kg-day)-1
This risk calculation also yields a unitless value. The acceptable individual
excess lifetime cancer risk range established by the USEPA [1] is 10-4 to 10-6
(one-in-ten thousand to one-in-one million excess cancer risk). Many state
regulatory agencies within the United States have established acceptable risk
target levels within this range.
25
Chapter 1
26
Chapter 1
Emergency Response
Emergency
Response
Needed?
Yes
No
Exceed
Satisfy
TIER 1 ASSESSMENT
Compare contamination to generic RBSLs
Yes
No
Corrective
action
feasible?
TIER 2 ASSESSMENT
Refine study, generate site-specific SSTLs
Compare contamination to SSTLs
Exceed
Yes
Satisfy
No
Corrective
action
feasible?
Exceed
Final Corrective
Action
TIER 3 ASSESSMENT
Further refine study and SSTLs
Compare contamination to SSTLs
Satisfy
NO FURTHER ACTION
27
Chapter 1
pose a significant risk to human health or the environment and that no remedial
action is necessary. If site concentrations exceed Tier 1 levels, the site manager
generally has the option of remediating the site to Tier 1 levels or, alternatively,
progressing to a more data and labor intensive Tier 2 or even Tier 3 analysis.
Tiers 2 and 3
Tier 2 and Tier 3 analyses generally require increasingly sophisticated levels of
data collection and analysis, which in turn result in increased costs. The tradeoff for these increased costs will generally lie in lower remediation and overall
project costs, because the cleanup goals defined by a Tier 2 or 3 analysis are
likely to be higher than Tier 1 levels, and thus less costly to achieve. The
cleanup goals of the Tier 2 and 3 analyses are generally higher than the Tier 1
analysis because the generic assumptions used in the Tier 1 levels are replaced
with more relevant site-specific assumptions and data. They are not higher
because they are less protective of human health or the environment. In fact, all
three tiers of risk analysis provide an equal level of health protection. Tier 2 and
3 risk-based concentrations are often referred to as Site-Specific Target Levels
(SSTLs). ASTM [3,4] defines an SSTL as a risk-based remedial action target
level for chemical(s) of concern developed for a particular site under the Tier 2
and Tier 3 evaluations.
Upon completion of each tier, the site manager reviews the results and
recommendations, and decides if the cost of conducting the additional sitespecific analyses is warranted. Using the tiered approach, an E&P site manager
has the flexibility to forego the detailed risk characterization effort of a sitespecific Tier 2 or 3 analysis and proceed directly to site actions that generally
involve meeting conservatively low, generic site cleanup goals. In some cases,
this approach may be the more cost-effective and more prudent site management
decision.
Implementation of a Tiered Risk Approach
The development of tiered approaches for the risk-based analysis of sites was
based on the premise that there are situations where conducting a detailed risk
analysis may require more effort and time than immediate implementation of
site remedial actions. For this reason, after every tier of risk analysis, the site
manager must perform a cost/benefit evaluation to determine if it makes sense to
proceed to the next level of risk analysis. Only if a clear benefit exists would
the decision to move forward be made. For example, because the Tier 1
assessment is often based upon conservatively low, generic site cleanup goals,
the extent of a site remedial action may be larger (and more expensive) than
might be required if a more detailed site-specific Tier 2 analysis were
conducted. However, additional time and expense will be incurred to complete
the Tier 2 analysis. At this point the site manager must evaluate the potential
reduction in site remedial costs that may be realized by conducting the Tier 2
28
Chapter 1
analysis and compare that reduction to the additional cost of conducting the risk
analysis. If the potential savings outweigh the potential cost, it would be in the
managers best interest to move forward with the Tier 2 analysis. In some cases,
it is not the cost that drives the decision, but the schedule. If the time required to
conduct the next tier of risk analysis is not acceptable to regulatory agencies or
the public, then the decision to proceed with site remediation is essentially
made.
The decision to use the tiered risk-based strategies for site management is
usually dictated by the nature of the site contamination and the complexity of
the site conditions. However, it may also be dictated by the governing
regulatory agency, which may or may not accept the use of a tiered approach. At
most E&P sites, it is likely that a tiered risk-based decision-making strategy will
be the approach of choice. This is because E&P sites generally involve a known
and very limited number of chemicals of potential concern (e.g., crude oil, gas
condensates, and selected additives), and they have relatively small and simple
operational footprints. Consequently, the lower tiers of risk analysis will often
provide the most cost-effective site management approach.
Role of Generic Site Cleanup Criteria
in the Risk-Based Decision-Making Process
As described in Chapter 2, most regulatory pro grams, including those with
jurisdiction over E&P sites, have historically incorporated cleanup criteria that
are not explicitly health risk-based levels. Instead, generic criteria have often
been used, such as the commonly applied 1% TPH management level. For
practical purposes, these generic criteria may be used as Tier 1 screening level
criteria in a pseudo risk-based decision-making process. However, it should be
recognized that those generic criteria that are not risk-based may or may not be
protective of human health and the environment. One of the goals of a recent
joint-industry research project sponsored by the Petroleum Environmental
Research Forum was to derive generic risk-based screening criteria for E&P
sites that could be compared to existing, non risk-based criteria currently used
for E&P site management. See Chapter 8 for more information on RBSLs for
TPH.
29
Chapter 1
days
year
- 6 kg
EF x ED
RfD o
where:
THQ
BW
ATn
EF
ED
IRsoil
RfDo
RAFd
RAFo
SA
M
=
=
=
=
=
=
=
=
=
=
=
30
Chapter 1
31
Chapter 1
sufficient to merely assume that decisions made and/or Tier 1 RBSLs derived
for programs developed in other parts of the world are appropriate for a new
application in a new location. Country and/or region-specific information is
necessary because living and working conditions in the Netherlands, Canada, or
the United States may be quite different than in countries like Nigeria or
Thailand. For example, some differences that may affect exposures include the
length of time that people typically live in the same house, whether groundwater
is consumed and how much is consumed, and how many days per year a worker
spends at his job. The Dutch do not include exposure to groundwater in their
risk program because their citizens do not drink groundwater. In the United
States, consumption of groundwater is common and this pathway is an integral
part of risk evaluation.
A brief summary of some of the international risk-based decision-making
programs and Tier 1 RBSLs used in these programs is provided below.
Dutch National Institute of Public Health and the Environment (RIVM)
In 1994, the Dutch RIVM (National Institute of Public Health and the
Environment) published soil and groundwater Target and Intervention Values
[14]. Intervention Values for soil are defined as the concentrations above which
soils are considered to be seriously contaminated, and are applicable to all sites
exceeding 25 m3 in size. Target Values indicate the soil quality levels ultimately
aimed for and are usually based on background chemical concentrations in soil
in the Netherlands. In applying these values to individual sites, it is generally
concluded that further site investigation is required if the soil concentration of a
given chemical exc eeds this criterion:
Intervention Value + Target Value
2
The target values are derived based on human health toxicity information for
each chemical of concern and residential exposure estimates are derived using
the Dutch CSOIL model. Several different residential exposure pathways are
considered including ingestion of soil, ingestion of crops grown in impacted
soil, and inhalation of indoor air. In selecting the overall exposure pathway of
concern, RIVM derives serious soil contamination concentrations (SCCs) for
each of the exposure pathways and then selects the most conservative (i.e., the
lowest) concentration to propose as the overall Intervention Value for that
chemical. A tiered risk approach is not used and commercial exposure scenarios
are not considered in the Netherlands.
Canadian Council of Ministers of the Environment (CCME)
In 1991, CCME published Interim Canadian Environmental Quality Criteria for
Contaminated Sites [15] and, in 1996, released a follow-up report, A Protocol
32
Chapter 1
for the Derivation of Environmental and Human Health Soil Quality Guidelines
[16], to derive Soil Quality Guidelines (SQGs) that will replace the 1991
criteria. The Canadian protocol considers potential exposures to both human
and ecological receptors for given land uses. SQGs are derived based on
exposure scenarios for agricultural, residential/parkland, commercial, and
industrial land uses for jurisdictions in Canada. For each of the four land uses,
CCME selects final generic SQGs for chemicals in soil based on the lowest
value generated by either of the environmental or human health approaches.
The protocol also identifies indirect routes of exposure including consumption
of meat, milk, and produce from agriculture land, and consumption of
homegrown produce from residential land.
In December 2000, the CCME published Canada-Wide Standards for Petroleum
Hydrocarbons (PHCs) in Soil: Scientific Rationale [17] to address different
types of hydrocarbon fractions. Following the approach described by the United
States Total Petroleum Hydrocarbon Criteria Working Group [18], the CCME
has defined four broad hydrocarbon fractions (F) including F1: C6 to C10; F2:
>C10 to C16; F3: >C16 to C34; and F4: C 34+. Aliphatic and aromatic sub-fractions
are handled separately. With respect to management of PHCs, additional factors
considered in CCMEs Tier 1 level include: ignition hazard, odor and
appearance, effects on buried infrastructure, formation of non-aqueous phase
liquids, and socioeconomics and technological capabilities.
New Zealand Ministry for the Environment
In 1999, New Zealands Ministry for the Environment published Guidelines for
Assessing and Managing Petroleum Contaminated Sites in New Zealand [19].
The Ministry presents a tiered risk-based approach to site assessment and
development of soil and groundwater acceptance criteria for petroleum
hydrocarbon contaminated sites. Tier 1 acceptance criteria are summarized in
look-up tables and site-specific criteria used to develop Tier 2 and 3 criteria are
outlined. Tier 1 acceptance criteria have been developed for a wide range of
possible site characteristics rather than a single generic scenario. For example,
the Tier 1 criteria include consideration of various land uses, such as
agricultural, residential, commercial/industrial, and maintenance workers . Also
considered are soil type (8 soil profiles are represented), depth of soil
contamination, e.g., <1 meter, 1-4 meters, and >4 meters, and depth of
groundwater from ground surface, such as 2-4 meters, 4-8 meters, and >8
meters. New Zealand specifies specific compounds of concern for each type of
refined product for use in deriving Tier 1 soil acceptance criteria.
References
1.
33
Chapter 1
2.
3.
American Society for Testing and Materials (ASTM), Standard Guide for
Risk-Based Corrective Action Applied at Petroleum Release Sites, ASTM
E1739-95, November 1995.
4.
American Society for Testing and Materials (ASTM), Standard Guide for
Risk-Based Corrective Action, ASTM E2081-00, April 2000.
5.
6.
7.
8.
9.
34
Chapter 1
35
Chapter 2
Introduction
This chapter reviews the technical basis for the 1% total petroleum hydrocarbon
(TPH) management level for land management of exploration and production
(E&P) wastes as recommended by the American Petroleum Institute (API) and
provided in state regulations within the United States (e.g., Louisiana (LA) 29b
and Texas (TX) Rule 91) and practice. The limitations of the 1% TPH
management level are discussed and the need to develop risk-based management
levels based on site-specific information is introduced.
Crude oil from a producing formation and diesel added to drilling mud are the
chief sources of petroleum hydrocarbons in E&P wastes [1]. Gross analysis of
petroleum hydrocarbons within a waste may be reported as oil and grease
(O&G), TPH, or total organic carbon (TOC). O&G is determined by
gravimetric analysis of a solvent extract, TPH is determined by infrared analysis
of a solvent extract, and TOC is analyzed by wet oxidation of a sample. O&G
was the predominant analytical method used to measure petroleum hydrocarbon
content of E&P wastes prior to the 1990s, with TPH becoming the predominant
analytical method since that time.
Initially, a 1% TPH concentration in soil was used as a guidance value for E&P
wastes that were land-managed. The 1% TPH guidance value was developed
based on two criteria: the first was the impact of hydrocarbons on plant life and
the second was hydrocarbon mobility to groundwater. Over time, certain United
States regulatory agencies adopted the 1% TPH guidance value as a regulatory
limit for the hydrocarbon content of land-disposed E&P wastes. At sites where
oily wastes having TPH concentrations greater than 1% have been landdisposed, the regulatory limit of 1% is sometimes used as a soil cleanup
standard. In the paragraphs that follow, the terms guidance value, regulatory
limit, and cleanup standard are used where appropriate; the term
management level is used in instances where any of these terms may be
appropriate.
36
Chapter 2
37
Chapter 2
Comments
Non-Sensitive
Sensitive
Land Treatment of NOW
Was 250 mg/kg
Site Dependent
Railroad Commission Rule 91
Site-by-Site Basis
38
Chapter 2
The damage that oil does is due mostly to the prevention of the plant
from obtaining sufficient moisture and air and from ramifying its roots:
very little is due to toxicity, as such.
Thus, the development of the 1% TPH guidance level in soil was originally
based on the issue of toxicity to plants. If the amount of petroleum hydrocarbon
in soil is kept at or below this level, and there are acceptable salt and pH levels,
there should not be significant reductions in plant yield, thus meeting landowner
requirements.
More recent studies confirm the earlier studies on plant growth and germination
for heavy or weathered crude oil. However, light or fresher crude oil may be
found to require more stringent guidance. Chaineau et al. [21] found that the
phytoavailability of complex mixtures of hydrocarbons that have low octanolwater partition coefficients (Kow) is negligible, even when the soil petroleum
hydrocarbon content is as high as 1%. Phytotoxicity was found to be greater for
low molecular weight and aromatic hydrocarbons and varied greatly with
hydrocarbon concentration in the soil and plant species [21]. Salanitro et al.
[22] found a similar molecular weight relationship. Residual TPH in which
germination was not affected (<4 to 27% reduction) in bioremediated soils (10
months treatment) varied from 7,000 to 10,000, from 8,200 to 8,600, and from
39
Chapter 2
1,000 to 1,200 mg/kg for the heavy, medium, and light oily soils, respectively
[22].
Current understanding of toxicity issues suggests that soil toxicity
considerations may expand beyond plant toxicity alone. Other potential
ecotoxicity issues are now being examined. For example, recent studies have
examined earthworms [23] as well as other soil invertebrates [24].
Petroleum Hydrocarbon Impact on Groundwater Resources
If a sufficient quantity of hydrocarbon is released to a soil, the hydrocarbon will
migrate through the soil as a non-aqueous phase liquid (NAPL). As the NAPL
moves, some will remain in the pore spaces of the soil, until there is no longer
sufficient volume of NAPL remaining for migration to occur. At this point, the
NAPL in the soil is said to be at residual saturation. The 1% soil TPH guidance
value was selected to be below the minimum level required for hydrocarbon
mobility (or less than the residual saturation). Therefore, the guidance value of
1% TPH in soil prevents movement of NAPL toward groundwater receptors.
Researchers have confirmed that hydrocarbon migration is not a problem at low
percentages of hydrocarbons in soil. Raymond et al. [25] found that by adding
approximately 2% oil to the top 15 centimeters of soil, 99% remained within the
top 20 centimeters after 1 year. When hydrocarbon loading rates of 3 to 13%
per year were added, no significant oil migration was found below the zone of
incorporation [25]. Brost and DeVaull [26] tried to determine a conservative
NAPL concentration in unsaturated soil below which NAPL would be
immobile. Unsaturated soil samples were saturated with hydrocarbons and then
allowed to drain. The amount of residual hydrocarbon remaining in the soil
pores was then quantified. Brost and DeVaull determined the residual saturation
for middle distillates and fuel oils to range from 0.8 to 5.0% [26]. The
variability is believed to be attributable to experimental method variability,
variation in soil type, unique chemical properties and measurement differences.
When oil enters the soil as a NAPL there is natural separation of the
hydrocarbon constituents due to exposure of the NAPL to the solid phase, vapor
phase, and water phase within the soil. The higher molecular weight compounds
are generally less mobile and stay near the source location, while the lighter
weight compounds migrate deeper into the subsurface because of greater
aqueous solubility [20]. Biodegradation of these compounds may play a role in
preventing transport to groundwater. Webster and Loehr [27] studied the rate of
hydrocarbon release for six soils containing petroleum hydrocarbons. Their
work indicated soil hydrocarbons containing slightly to moderately weathered
diesel range organics were more available for release when compared to
weathered crude oil [27].
40
Chapter 2
41
Chapter 2
Summary
The 1% TPH management level used by Texas, Louisiana, and other regulators
has been shown to be protective of groundwater resources and plant life. Also,
the 1% TPH management level was shown to be achievable, specifically through
bioremediation.
Although the 1% TPH management level may be adequate in many cases, riskbased cleanup standards are still needed to address the human health risks of
particular hydrocarbon mixtures. The specific components of crude oil must be
identified and the potential risks evaluated for those sites at which human
exposures may be of concern. For this reason, within the last 10 years, technical
methods have been developed to determine soil and groundwater corrective
action cleanup criteria using a site-specific and human health risk-based
approach. These methods are summarized in this book, and crude oil
composition data are provided to aid in the risk evaluation. Risks to ecological
receptors (other than plants) may also need to be considered, but the state of the
science is not as advanced as that for human health risk assessment.
42
Chapter 2
References
1.
Deuel, Jr., L.E., and Holliday, G.H., Soil Remediation for the Petroleum
Extraction In dustry, PennWell Books, Tulsa, Oklahoma (1994).
2.
3.
Miller, R.W., and Honarvar, S., Effect of Drilling Fluid Components and
Mixtures on Plants and Soils, conference proceedings: The Environmental
Aspects of Chemical Use in Well Drilling Operations, Environmental
Protection Agency, EPA/1-75-004 (1975).
4.
5.
6.
7.
Wascom, C.D., Oilfield Pit Regulations, A First for the Louisiana Oil &
Gas Industry, proceedings of a National Conference on Drilling Muds,
University of Oklahoma, Norman, May 29-30, 1986.
8.
9.
10. McMillen, S.J., Magaw, R. I., Kerr, J. M., Sweeney, R. E., Nakles, D. V.,
and Geiger, S.C., A New Risk-Based Approach to Establish Cleanup
Levels for Total Petroleum Hydrocarbons, Proceedings of the 6th
International Petroleum Environmental Conference, Houston, Texas,
pp. 438-459, (2000).
11. Colorado BLM Policy for the On-Site Remediation of RCRA Exempt,
Hydrocarbon Based, Oil and Gas Exploration & Production Waste and
Terrestrial Spills of Crude Oil, June 1997.
12. Colorado Oil & Gas Co mmissions (COGCC) Rules 907 and 908 (1995),
and Revisions to the 900 Series Rule (1997).
43
Chapter 2
44
Chapter 2
45
Chapter 3
Introduction
One of the basic principles of risk-based decision-making is that the decisionmaker must have a good understanding of the composition of the material of
concern. Therefore, knowledge of the chemical and physical characteristics of
crude oils and condensates is required for the effective application of risk-based
decision-making at exploration and production sites. A summary of the
chemical and physical properties of both condensates and crude oils is presented
in this chapter. Crude oils are defined as mixtures of hydrocarbons that exist in
the liquid phase in underground reservoirs and that remain in the liquid phase at
atmospheric pressure. Condensates are mixtures of hydrocarbons that are in the
vapor phase under reservoir pressures and temperatures but become liquid under
atmospheric conditions.
46
Chapter 3
Saturates:
(alkanes or paraffins)
Unsaturates:
Olefins:
n-Alkane:
H
H
H
H | H | H | H
| C | C | C |
C | C | C | C
| H | H | H |
H
H
H
H
H
|
C
H
H
H
H | H | H | H
| C | C | C |
C | C | C | C
| H | H | H |
H
H
H
H
|
C
|
H H H
Aromatics:
Isoalkane:
H
H
C = C
C
|
C
H
H
C
||
C
C
|
H
Double
Carbon
Bond
H
Cycloalkane:
H
H
H
H
C
C
C
C
C
H H
47
Chapter 3
Petroleum
Hydrocarbons
Saturates (also
known as paraffins
or alkanes)
n-alkanes
(straight
chain)
isoalkanes
(branched
chain)
Unsaturates
cycloalkanes
or naphthenes
(cyclic)
Aromatics
Olefins
48
Chapter 3
Paraffinic
Pennsylvania
Rumania
Poland
Mid-continent
Baku
Naphthenic
Grozny
Borneo
Borneo
Japan
Aromatic
Mexico
Rumania
N. Calif
Gulf
Coast
Asphaltic
Weight Percent
25
50
17
8
100
The composition of 636 crude oils from around the world have been compared
by Tissot and Welte as shown in Figure 4 [7]. These data reveal that the
proportions of saturates, aromatics, resins, and asphaltenes can vary dramatically
among crude oils , with the majority of crude oils lying within a composition
envelope that is bounded in the following manner:
40 to 80% Saturates
15 to 40% Aromatics
0 to 20% Resins and Asphaltenes
Tissot and Welte found that 95% of the crude oils produced around the world
fell into this distribution pattern as shown in Figure 4.
49
Chapter 3
Aromatics 100%
95% Frequency
Distribution Envelope
World Oils
50%
100%
Saturates
50%
50%
100%
Resin + Asphaltenes
50
Chapter 3
Figure 5. The boiling point ranges and carbon ranges for six common crude oil
products [8].
Table 3. Composition of a 35API-gravity crude oil [1].
Molecular Size
Gasoline (C5-C10)
Kerosene (C8-C12)
Diesel Fuel (C13-C17)
Heavy Gas Oil (C19-C25)
Lubricating Oil (C20-C45)
Residuum (>C40 )
Total
Volume Percent
27
13
12
10
20
18
100
51
Chapter 3
Widuri crude oil from Sumatra is dominated by normal alkanes or paraffins that
produce a picket fence type pattern in the chromatograph, which is typical of
waxy crude oils. On the other hand, the San Joaquin Valley (SJV) crude oil
from California is dominated by a hump or unresolved complex mixture of
hydro carbons that is difficult for a gas chromatograph to separate. This hump is
indicative of the prior biodegradation of hydrocarbons that occurred in the oil
reservoir and is a common characteristic of many heavy crude oils.
Gasoline
52
Chapter 3
Diesel
53
Chapter 3
many condensates that come directly from wells tend to have a tail of heavier
hydrocarbons, while condensates that come from natural gas processing plants
or from condensation in pipelines have a much narrower range. This is because
some amount of processing has occurred, and these types of condensates will
have carbon ranges similar to refined gasoline.
Gas chromatography fingerprints of the saturated and aromatic hydro carbon
fractions of two condensates are shown in Figure 7. These fingerprints illustrate
the large degree of variability that can exist for these hydrocarbon mixtures. In
particular, it is clear that Condensate A encompasses a much broader range of
hydro carbons than does Condensate B. Also, the ratio of the saturated
hydrocarbons to the aromatic hydrocarbons is quite different for these two
condensates, increasing from 3.2 for Condensate B to 5.8 for Condensate A.
Condensate B
Condensate A
Saturate
Saturate
* * *
Aromatic
Aromatic
*
*
* Internal Standard
Compounds of Concern in
Crude Oils and Gas Condensates
Following crude oil or condensate releases to the environment, regulators
usually require that soils and/or groundwater be analyzed for TPH. Some
jurisdictions may also require analysis for other compounds of concern (COCs).
These COCs might include specific volatile hydrocarbons such as benzene,
toluene, ethylbenzene, and xylenes; the 16 semi-volatile polyaromatic
hydrocarbons that are on the EPAs priority pollutant list; and some metals. The
amount of each of these COCs in crude oils and condensates as well as the
implications of their presence on risk-based decision-making for E&P sites are
discussed in detail in Chapters 10 through 12.
54
Chapter 3
141 .5
Specific Gravity @ 60 F
131.5
Pour Point: The pour point is the temperature below which a crude oil
will not flow in a horizontal tube [1]. The pour point for most oils
arises from the precipitation of wax such that a pasty, plastic mass of
interlocking crystals is formed. Wax-free oils have pour points that are
dependent upon viscosity only and will tend to thicken to glassy
materials as the temperature is reduced and the viscosity increases.
Some waxy crude oils may be solid at temperatures as high as 110F
(43C).
Crude Oil
Crude oil composition varies greatly and the differences in composition are
reflected in the API gravity values for crude oils. For example, heavy API
gravity oils (<20API) have higher concentrations of asphaltenes and resins than
55
Chapter 3
do light API gravity oils (>30API). Crude oils produced in the United States
have API gravity values that range from approximately 7 to 50 [6].
Crude oil is less dense than water with a specific gravity ranging from 0.85 to
0.98 (as compared to 1.0 for water). However, because of the large differences
in composition among the various crude oils, the precise specific gravity of the
crudes can vary substantially.
Crude oil also tends to be a viscous liquid at surface temperatures and pressures.
Surface viscosity values range from 1.9 to 19,400 centistokes [3]. Pour point
values for crude oils range from 70 to 110F [1]. Therefore, some crude oils
may be solid at typical seasonal fall and spring temperatures in the United
States. The viscosity and pour point are important because they imply that many
crude oils are not fluid enough to rapidly percolate through soil.
Crude oil is sparingly soluble in water, with solubility increasing with API
gravity. For example, a crude oil with an API gravity of 11 had a total
solubility in water of 3.5 mg/L at 25C (77F), whereas an oil with an API
gravity of 28 had a solubility of 65 mg/L [10]. However, total solubility is
dependent on temperature and the composition of the crude oil.
Condensates
Extensive physical property data are not currently available for condensates.
However, in broad terms, these hydrocarbon mixtures generally exhibit an API
gravity of greater than 45. This suggests that they are not extremely viscous at
normal ambient temperatures and that they are relatively volatile and soluble in
water. Composition data [11] for four condensates revealed that high molecular
weight alkanes can be present. The presence of these alkanes would have a
tendency to increase both density (i.e., decrease API gravity) and viscosity and
decrease both solubility and volatility of the hydrocarbon mixture.
Summary
The composition of crude oils and condensates can vary greatly as reflected by
their class compositions, carbon ranges , and other properties such as API
gravity. Composition may affect fate and transport in the environment, and
therefore can impact risk-based decision-making. While it is possible to
generalize as to crude oil and condensate content and properties, understanding
the specific composition of oils and condensates is required before
implementing risk-based decision-making at E&P sites. For this reason, the
Petroleum Environmental Research Forum Project 97-08 embarked on analyzing
a large number of crude oils and condensates that would be representative of
those produced around the world. The data from this projectare presented in the
following chapters.
56
Chapter 3
References
1.
2.
3.
4.
5.
6.
7.
Tissot, B.P., and Welte, D.H., Petroleum Formation and Occurrence, Berlin:
Springer-Verlag (1978).
8.
9.
57
Chapter 4
Introduction
The Total Petroleum Hydrocarbon Criteria Working Group (TPHCWG)
convened in 1993 to develop scientifically defensible information for
establishing soil cleanup levels that are protective of human health at petroleum
release sites [1-5]. The impetus for the formation of this group was the large
disparity in cleanup standards for total petroleum hydrocarbons (TPH) in soil,
and the recognition that these regulatory standards were not based on risk to
human health. Active participants in this effort were the Air Force, Exxon,
Shell, Chevron, the American Petroleum Institute (API), the Association of
American Railroads, several state governments (Washington, Texas, Colorado,
Hawaii, Louisiana, New Mexico, Massachusetts) of the United States , the
United States Environmental Protection Agency (USEPA), the Department of
Defense, as well as private consulting companies including EA Engineering
Science and Technology and Menzie-Cura & Associates, Inc.
The use of risk-based decision-making for petroleum mixtures is complicated
due to the fact that petroleum consists of several thousand individual
hydrocarbons and other compounds, each with a unique set of physical and
chemical characteristics including volatility and solubility. Only about 250 of
these compounds have been specifically identified, and it is impossible to
analyze all of the specific constituents in most petroleum products or crude oils.
In response to this difficulty, the TPHCWG chose to use a fractionation
approach to analyze for TPH and coupled this data with the standard risk
assessment approach for deriving Tier 1 risk-based screening levels (RBSLs)
described by the American Society for Testing and Materials (ASTM) [6].
Similar approaches for deriving TPH RBSLs or cleanup levels have been
adopted by the States of Massachusetts, Texas, and Louisiana and incorporated
into their environmental management programs for downstream sites [7-9].
58
Chapter 4
The TPHCWG summarized their findings in five volumes [1-5]. This chapter
provides a brief overview of their major findings, and presents new data on the
composition of crude oils and condensates. Before considering the new
analytical approach developed by the TPHCWG, the limitations of older TPH
analytical methods are described.
59
Chapter 4
60
Chapter 4
Isomers
1 Each
2
3
5
9
18
35
Size
C10
C11
C12
C13
C15
C18
C25
Isomers
75
159
355
802
4,347
60,523
36,797,588
61
Chapter 4
Aromatics
H
|
C
HC
CH
HC
CH
Aliphatics
H H H
H H H
H H
|
|
|
|
|
|
|
|
H C C C C C C C C H
|
|
|
|
|
|
|
|
H H H
H H H H H
C
|
H C H
|
H
>C5 -C6
>C6 -C8
>C8-C10
>C10-C12
>C12-C16
>C16-C35
62
Chapter 4
USEPA Method 8260 is required to analyze for benzene (the only hydrocarbon
in the aromatic C6-C7 fraction), and toluene (the only hydrocarbon in the
aromatic C7-C8 fraction), rather than using the TPHCWG analytical method.
The TPHCWG analytical method is provided in the Appendix.
The Basis for Defining 13 TPH Fractions
The EC fractions were identified by selecting groups of hydrocarbons that have
similar fate and transport properties, such as solubility and vapor pressure. This
was done because of the important role that fate and transport play in
determining the exposure of a receptor to a site contaminant. For example,
highly soluble petroleum compounds are more likely to migrate to groundwater
and represent potential risk to humans via the consumption of drinking water.
By choosing fate and transport criteria for the definition of the fractions, the
TPHCWG ensured that the risk assessment would properly capture the fraction
of the hydrocarbon mixture that would be present at the point of exposure for the
variety of exposure pathway-receptor combinations that might be present at a
site.
Aliphatic and aromatic hydrocarbon fractions were considered separately
because their solubility and other fate and transport characteristics are so
dramatically different (Table 2). Within each of these groups, the major
differences in fate and transport properties were related to the EC numbers of
the compounds. As shown in Table 2 and Figure 3, the properties of the
compounds (i.e., solubility, vapor pressure) and their modeled environmental
behavior [i.e., coefficient for partitioning to organic carbon (Koc) from soil to
water (LF sw)] change by an order-of-magnitude or more between the different
EC number fractions.
63
Chapter 4
Aliphatic Fractions
C5-C6
>C6-C8
>C8-C10
>C10-C12
>C12-C16
>C16-C35
Aromatic Fractions
C6-C7 (Benzene)
>C7-C8 (Toluene)
>C8-C10
>C10-C12
>C12-C16
>C16-C21
>C21-C35
Solubility
(mg/L)
Vapor
Pressure
(Atm)
Log Koc
H
(cm 3/cm 3)
LFsw
(mg/L)/
(mg/kg)
3.6E+01
5.4E+00
4.3E-01
3.4E-02
7.6E-04
1.5E-06
3.5E-01
6.3E-02
6.3E-03
6.3E-04
4.8E-05
1.1E-06
3.0E+00
3.6E+00
4.8E+00
5.9E+00
6.7E+00
8.6E+00
3.40E+01
5.10E+01
8.20E+01
1.30E+02
5.40E+02
1.10E+02
6.25E-03
1.73E-03
2.51E-04
3.26E-05
1.64E-06
8.26E-09
1.8E+03
5.2E+02
6.5E+01
2.5E+01
5.8E+00
5.1E-01
2.9E-02
1.3E-01
3.8E-02
6.3E-03
6.3E-04
4.8E-05
7.6E-06
1.6E-08
1.9E+00
2.4E+00
2.9E+00
3.2E+00
3.8E+00
4.2E+00
5.1E+00
2.25E-01
2.70-01
4.90-01
1.40-01
5.40E-02
1.20E-02
8.20E-05
9.00E-02
3.37E-02
5.16E-03
3.28E-03
1.64E-03
5.21E-04
6.56E-05
Figure 3. Solubility vs. equivalent carbon number for aromatics and aliphatics
[3].
64
Chapter 4
Toxicity criteria were developed for each fraction as shown in Table 3. The
methodology used by the USEPA and the TPHCWG for developing the
reference doses (RfDs) and reference concentrations (RfCs) is described in
detail in Chapter 7.
Table 3. Toxicity criteria for TPH fractions [4].
Equivalent Carbon
Number Range
Aliphatic
<C6
>C6-C8
Aromatic
C6-C7 (Benzene)
>C7-C8 (Toluene)
Aliphatic
>C8-C1 0
>C10-C12
>C12-C16
Aromatic
>C8-C1 0
>C10-C12
>C12-C16
Aliphatic
>C16-C35
Aromatic
>C16-C35
Oral RfD
(mg/kg/day)
5.0
Inhalation RfC
(mg/m3)
18.4
0.2
0.4
0.1
1.0
0.04
0.2
2.0
NA
0.03
NA
NA = Not available
Adapting the TPHCWG Analytical Methodology to Crude Oils
The original TPHCWG approach does not include hydrocarbons greater than
carbon number 28 (C28), but has been modified to include up to C35 [3]. This is
appropriate for most refined products, such as gasoline and diesel, as well as for
condensates. However, some crude oils with very low API gravity values may
contain as much as 50 to 60% hydrocarbons >C35. Therefore, the TPHCWG
analytical methodology was further modified so that hydrocarbons up to C44
could be fractionated and detected by GC (as described in Chapter 5). Figure 4
presents the data for 15 different crude oil samples plus a diesel sample using
the extended GC analytical method. From this figure, it can be seen that
approximately 45 to 80% of the hydrocarbons in crude oil can be detected using
a gas chromatograph (i.e., EC numbers from C6 to C44).
65
Chapter 4
100
C6GC
to -C44
%
C6 to C44
<C6
>C45
% Detected by GC
90
80
70
60
50
40
30
20
10
0
5 6 7 8 9 10 11 12 13 14 15
Different Crude Oils
Diesel
Figure 4. Mass balance obtained for crude oils using a modified TPHCWG
analytical method and distillation data.
The remaining hydrocarbons fall into two groups: those <C6 or those >C44.
These ranges are typically quantified using distillation to determine the entire
composition of a crude oil. The hydrocarbon fraction with carbon numbers
greater than C44 is sometimes called the vacuum residuum, because it contains
the compounds remaining after the vacuum distillation of crude oil. The less
than C6 fraction is lighter than gasoline and is predominately the component of
natural gas. The amounts of hydrocarbons <C6 and >C44 (obtained from
distillation) for each crude oil are also shown in Figure 4. The addition of all
three molecular weight ranges accounts for more than 85% of the compounds
within crude oils. At the same time, greater than 95% of diesel oil can be
detected by the TPHCWG analytical method alone, further reinforcing that the
TPHCWG method adequately quantifies refined products such as diesel.
There are no standard USEPA analytical methods available for quantifying
either the <C6 or the >C44 fraction (vacuum residuum). Because the <C6 fraction
is generally lost to volatilization after a crude oil release, it is likely to be
unnecessary to quantify this fraction for assessing potential health risks for soil
exposures. On the other hand, the fate and transport characteristics of the >C44
fraction or vacuum residuum (see Table 4) indicate that it may remain in soils
even after extensive weathering of an oil has occurred.
The vacuum residuum fraction of a crude oil is comprised of very large
molecules (those boiling above 600C) that are not well characterized as to their
compositional make up, but it is known to contain a mixture of aliphatics,
aromatics, metals, and asphaltenes. Because of the complex nature, limited
66
Chapter 4
mobility, and the small amount of published toxicity data on this fraction, a
decision was made to evaluate it as a single fraction, rather than trying to
separate it into its aliphatic and aromatic components. The amount of vacuum
residuum in 800 crude oils from the United States is shown in Figure 5 [14].
80
70
60
50
40
30
20
10
0
0
10
20
30
40
50
60
70
Figure 5. The yield of vacuum residuum in 800 crude oils produced in the
United States [14].
To determine the mass of vacuum residuum in a crude oil, three approaches can
be used: (1) using the known amount of vacuum residuum present in a crude oil
as determined by distillation; (2) estimating the amount of vacuum residuum
from the API gravity; and (3) assuming that all material not accounted for by
GC is vacuum residuum. The first method is probably the most reliable, but
distillation data are not always available and may be costly to obtain. The
second method provides a rough approximation since it uses the slope of the line
that is fit through the data shown in Figure 5, so that:
% Yield of Vacuum Residuum = -1.253(API Gravity) + 69.32
This equation has a R2 value of 0.66. The third method is probably the least
reliable, but may be an acceptable approach if neither distillation nor API
gravity data are available.
67
Chapter 4
Aromatics
Aliphatics
H
|
C
HC
CH
HC
CH
H H H H H H H H
|
|
|
|
|
|
|
|
H CCCCCCCC H
|
|
|
|
|
|
|
|
H H H H H H H H
C
|
H C H
|
H
>C6-C7 (benzene)
>C7-C8 (toluene)
>C8-C10
>C10-C12
>C12-C16
>C16-C21
>C21-C44
>C5-C6
>C6-C8
>C8-C10
>C10-C12
>C12-C16
>C16-C44
>C44
Figure 6. Aliphatic and aromatic carbon number fractions for the assessment of
risk associated with crude oil TPH (highlighted fractions are different than
fractions of TPHCWG).
68
Chapter 4
The fate and transport chara cteristics for the C35-C44 and the >C44 fractions are
presented in Table 4. The toxicity characteristics were evaluated for these
fractions as described in detail in Chapter 7. Because of the fate and transport
characteristics of the C44 fraction, the only exposure pathway of concern will be
direct contact with surface soil.
Table 4. Fate and transport characteristics of additional TPH fractions (based
on equivalent carbon number) [11].
Aliphatic Fractions
>C16-C44
Aromatic Fractions
>C21-C44
Vacuum Residuum
>C44
Vapor
Solubility Pressure koc
(mg/L)
(Atm)
(cm 3/g)
H
(cm 3/cm 3)
LFsw
(mg/L)/
(mg/kg)
1.3E-06
7.6E-06
1.00E+09
6.40E+03
8.26E-09
6.6E-03
4.4E-09
1.26E+05
6.80E-04
6.56E-05
1.0E-04
NA
5.01E+05
4.10E-08
1.65E-05
69
Chapter 4
800000
Concentration (mg/kg
700000
600000
500000
Diesel
Gasoline
400000
Vaseline
Crude oil
300000
200000
100000
>C
44
be
nz
en
e
>8
tol
-1
ue
0
n
e
Ar
om
>1
0ati
12
cs
Ar
om
>1
2ati
16
cs
Ar
om
>1
ati
6cs
21
Ar
om
ati
cs
>2
1
Ar
om
ati
cs
>6
-8
Al
iph
>8
ati
-1
cs
0
Al
iph
>1
a
0tic
12
s
Al
iph
>1
2-1
at
i
cs
6
Ali
ph
ati
cs
>1
6A
lip
ha
tic
s
TPH Fractions
70
Chapter 4
10
16
3 1
1
8
2
53
3
33
4
1
Figure 8. Number of crude oil samples (70 total) analyzed by the geographic
region from which they originated [11].
One of the goals of the PERF study was to analyze enough crude oils so that the
results would be representative of all the crude oil types produced around the
world. Overall, crude oils from around the world exhibit a wide range in
molecular composition. However, if crude oils are classified on the basis of
aliphatic, aromatic, and non-hydrocarbon abundance, then the range in
compositional variation is limited due to the similarity in organic matter types
forming crude oil and predictable compositional changes that occur during crude
oil alteration. To illustrate this point, Tissot and Welte (1978) plotted the
composition of 636 crude oils from around the world and showed that crude oil
compositions fall in a narrow band extending from thermally mature crude oils,
enriched in aliphatic hydrocarbons, to biodegraded crude oils, with low amounts
of aliphatic hydrocarbons [15]. Figure 9 shows the composition of 51 crude oils
and six oily soils from the PERF study overlain on the global distribution for
crude oils from the Tissot and Welte study, and indicates that the analyzed oils
adequately represent the worlds oils.
71
Chapter 4
Aromatics 100%
l
95% Frequency
Distribution Envelope
World Oils
50%
100%
Saturates
50%
50%
100%
Resin + Asphaltenes
72
Chapter 4
36,000
48,000
36,000
83,000
200,000
11,000
9,600
31,000
39,000
85,000
230,000
24.5
760
4,100
14,000
18,000
68
410
10,000
7,200
1,600
25
220,000
140,000
73,000
180,000
410,000
48,000
31,000
94,000
72,000
220,000
570,000
37,000
36,000
18,000
37,000
74,000
10,000
5,600
16,000
15,000
54,000
160,000
3
0
0
0
0
0
0
0
0
0
1
70
70
70
70
70
70
70
70
70
70
41
4,900
1,300
73
0.16 *
16,000
5,900
4,300
1,600
0
3
15
71
4,500
30
25,000
5,700
71
1,100
6,500
1.9
7.3
4,600
27,900
1,023
6,600
3
0
71
71
73
Chapter 4
218,231
154,296
74,117
69,778
44,936
60,577
19,837
15,633
6,089
3,707
56,000
85,000
569
ND
ND
24270
ND
ND
ND
ND
ND
670,394
230,430
130,000
210,000
200,000
117,399
29,000
43,000
28,000
20,000
ND
48,000
46,550
37,000
65,000
63,000
29,000
7,500
13,500
9,800
7,100
0
0
0
1
2
0
1
1
4
9
14
14
14
14
14
14
14
14
14
14
14
14
9,500
2,500
24,000
7,900
10
31,000
14,000
53,000
15,000
10
5,000
33,000
1,400
8,200
6,700
61,000
2,000
18,000
0
0
10
10
Summary
Most TPH analytical methods are not adequate for assessing potential risk to
human health because they are non-specific and provide little information on the
types of compounds present in terms of class or carbon range. The TPHCWG
developed a risk evaluation approach for complex petroleum hydrocarbon
products, based on an analytical method that divides the complex mixture into
several distinct carbon number fractions coupled with standard risk assessment
equations for assessing human exposure. With slight modifications, the
TPHCWG analytical method can be used to assess the hydrocarbon content of
crude oils. These modifications include extending the TPHCWG analytical
method to C44 and determining the fraction >C44 using one of three methods.
However, some regulatory agencies may opt to omit hydrocarbons >C44 in risk
assessments due to their lack of mobility in the environment.
TPH fraction data derived using the modified TPHCWG methodology can be
used to estimate potential non-carcinogenic human health risks. Since crude oils
vary widely in composition, 70 different crude oils were analyzed that were
representative of all the types of crude oils produced around the world. Data for
the equivalent carbon fractions and BTEX were presented. TPH RBSLs derived
using these data are presented in Chapter 8. Chapters 10 and 11 present more
74
Chapter 4
References
1.
2.
3.
Gustafson, S., Tell, S.G., and Orem, D., Selection of Representative Total
Petroleum Hydrocarbon (TPH) Fractions Based on Fate and Transport
Considerations, TPHCWG, Vol. 3, Amherst Scientific Publishers,
Amherst, Massachusetts (1997).
4.
5.
Vorhees, D., Gustafson, J., and Weisman, W., Human Health Risk-Based
Evaluation of Petroleum Contaminated Sites: Implementation of the
Working Group Approach, TPHCWG, Vol. 5, Amherst Scientific
Publishers, A mherst, Massachusetts (1999).
6.
ASTM (American Society for Testing and Materials), Standard Guide for
Corrective Action Applied at Petroleum Release Sites, ASTM E 1739-95,
November 1995.
7.
8.
9.
10. McMillen, S.J., Magaw, R.I., and Spence, L.R., Developing Risk-Based
Cleanup Standards for Soils in Nigeria, SPE Paper 66513, presented at the
SPE/EPA/DOE Environmental Conference, San Antonio, Texas (2001).
11. McMillen, S.J., Magaw, R.I., Kerr, J.M., Sweeney, R.E., Nakles, D.V., and
Geiger, S.C., A New Risk-Based Approach to Establish Cleanup Levels
75
Chapter 4
76
Chapter 5
Introduction
In evaluating the application of the TPHCWG analytical method (see Appendix)
to crude oil and crude oil impacted sites, several important factors affecting
method performance were considered. The factors investigated included soil
type, oil characteristics (unlike refined products, crude oils vary greatly),
extraction procedures, fractionation procedures (separation of the extract into
aliphatic and aromatic hydrocarbons), and chromatographic conditions. Each of
these factors is described in the paragraphs below.
Soils
Soil types investigated were s and, loam, and clay. Table 1 summarizes the soil
samples and their characteristics. The petroleum hydrocarbons that were added
to the soils included aged hydrocarbons (residual from historic releases) and
fresh oil added at the laboratory.
Table 1. Soil and hydrocarbon type used to evaluate the TPHCWG method.
Sample
Soil Type
Previous Oil
Lab Spike
(% by Weight)
Oil Type
(API)
1
2
3
4
5
6
Sand
Sand
Loam
Loam
Clay
Clay
2%
1.5%
4%
2%
3%
1%
38
11
15
34
25
22
Oils
Six different crude oils, in addition to the historic, aged hydrocarbons, were used
for the study. The American Petroleum Institute (API) gravity (o) of the oils
ranged from 11o to 38o. Oil was spiked into each soil sample to achieve a TPH
77
Chapter 5
concentration of between 1.5 and 4% oil by weight. Total recovery of oil was
calculated against an identical direct analysis of the oil. Heavier oils, with lower
API gravity values, tend to have greater proportions of >C28 material, while
some lighter oils, with higher API values, tend to have greater proportions of
paraffins (normal alkanes). These characteristics may bias the analysis of crude
oils using the current TPHCWG method.
Extraction
Four extraction procedures were investigated. If different extraction procedures
provided comparable results, then commercial laboratories would have greater
freedom in meeting a performance-based criteria method, and data users would
have greater opportunity to compare data from different sources.
Fractionation
Four different fractionation (separation of the extract into aliphatic and
aromatic) procedures were evaluated. Two procedures are provided in the
TPHCWG method based on modified USEPA Methods 3630 Silica Gel and
78
Chapter 5
3611 Alumina Separations [1]. The other two procedures are modifications of
USEPA Method 3630 [1], Silica Gel Separation.
The fractionation methods were evaluated in two ways. First, a laboratory
standard containing normal alkanes, benzene, toluene, ethylbenzene, and
xylenes (BTEX), and the 16 priority pollutant polyaromatic hydrocarbons
(PAHs) was prepared and fractionated by each procedure. Second, direct
analysis of non-fractionated oils and extracted soil residues were compared to
post-fractionation data to monitor mass balance (aliphatic/aromatic ratios) and
extraction recovery efficiency.
The TPHCWG provides two procedures for the separation of hydrocarbons into
aliphatic and aromatic class based extracts. These employ alumina or silica gel
chromatography. Both methods use small volumes for ease, expense, and to
alleviate the need to concentrate extracts. The TPHCWG procedures plus the
two additional procedures are summarized below.
USEPA Method 3630 [1] was modified to use 11 g silica gel as the
solid phase and pentane (to elute the aliphatics) and 1:1
pentane:methylene chloride (to elute the aromatics). This method
resulted in greater volumes of solvent being used that would require
extract concentration to maintain detection limits.
Chromatographic Conditions
Chromatographic conditions for the gas chromatography with flame ionization
detection (GC/GFID) described in the TPHCWG method were designed for use
with refined petroleum products such as gasoline and diesel. Thes e conditions
79
Chapter 5
may not be optimal for analysis of crude oils and were examined in this study.
Chromatographic conditions affecting the analysis of crude oil that were
investigated include loading capacity, effective carbon elution range, and mass
discrimination.
Results
This study was designed to evaluate the impact of sample processing on the
analysis of crude oil for TPHCWG method results. The critical elements in the
performance of this method were identified as extraction, fractionation of the
extract into aliphatics and aromatics, and GC/FID separation (chromatography).
Extraction
Table 2 compares the total recoverable petroleum hydrocarbon results for the six
soil types and four extraction techniques. Pentane does appear to be an adequate
solvent for extracting crude oils from soil, achieving recoveries similar to those
for methylene chloride. However, these results further demonstrate that TPH is
method defined, with each solvent and extraction procedure giving somewhat
different results. Laboratories should be flexible in their performance of the
method and evaluate potential bias by analyzing a matrix spike and matrix spike
duplicate with each analytical batch.
Table 2. Total recovered petroleum hydrocarbons (mg/kg).
Method
Concentration of
Added Oil in Soil
Vortex Pentane
Vortex Methylene
Chloride
Soxhlet Methylene
Chloride
Sonication
Methylene Chloride
1-Sand
30,000
2-Sand
15,000
Sample
3-Loam
4-Loam
40,000
20,000
5-Clay
40,000
6-Clay
20,000
24,700
23,200
8,160
7,500
21,700
33,700
15,800
16,400
30,600
43,000
7,250
11,200
20,800
7,170
18,100
15,100
41,100
14,400
NA
6,730
19,400
14,900
45,800
20,800
Fractionation
All four fractionation procedures were evaluated using a laboratory standard
containing normal alkanes, BTEX, and the 16 priority pollutant PAHs. As with
all fractionation procedures, it is imperative that the effectiveness of the
separation column be tested prior to any sample analysis. Several factors
influence the efficiency of a fractionation column to resolve compound classes.
They include activation/deactivation of the silica or alumina, type of solvents,
type of oil, and column overloading. Potential deactivation of the pre-packaged
80
Chapter 5
columns with moisture is a potential problem and they are therefore not
recommended.
For comparison purposes, GC/FID chromatograms for the TPHCWG and
Modified USEPA Method 3630 fractionation procedures are provided in Figures
1a and 1b, respectively. Figure 1a shows some compound class carryover
observed in the TPHCWG fractionation procedure which demonstrates the need
for the laboratory to monitor the fractionation efficiency carefully. When this
situation is encountered, the column must be reactivated before any samples are
analyzed.
F1 (aliphatic)
Response_
60000
55000
11199714.D\FID1A
50000
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Time
F2 (aromatic)
Response_
11199725.D\FID1A
40000
35000
Irregular
distribution of
PAHs indicates
problem
30000
25000
20000
15000
10000
5000
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Time
81
Chapter 5
F1 (aliphatic)
Response_
No aromatics
observed in F1
fraction
55000
50000
11199736.D\FID1A
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Time
F2 (aromatic)
Response_
11199747.D\FID1A
50000
45000
40000
35000
30000
25000
20000
15000
10000
5000
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
Time
Total
Aromatic
(mg/kg)
3,100
5,000
5,300
4,700
%
Aliphatic
81
70
67
70
82
Chapter 5
Method
TPHCWG Alumina
TPHCWG Silica
USEPA 3630 Silica
USEPA 3630 Silica HPLC
Total
Aliphatic
(mg/kg)
5,400
4,900
4,200
4,100
Total
Aromatic
(mg/kg)
3,700
4,000
4,100
4,000
%
Aliphatic
59
55
51
50
83
Chapter 5
Figure 2. Aliphatic (a) and Aromatic (b) fractions of high API gravity crude oil.
However, in biodegraded oils (Figure 3a, low API gravity) this aliphatic
hydrocarbon pattern may not be as evident due to the biodegradation of the nalkanes. In the case of the aromatic fraction, the high API gravity oil
demonstrates a characteristic pattern of one, two, and three ring aromatics
(Figure 2b); however, in the low API gravity oil the characteristic pattern is
degraded and replaced by an unresolved aromatic complex mixture (Figure 3b).
Identification of carryover in the low API gravity oil is more difficult than the
more common lighter crude oils, therefore careful analysis of quality control
samples is critical for the successful application of this procedure. For waxy
crude oils, the carryover of high molecular weight n-alkanes is a frequent
problem that can only be identified by evaluating the chromatograms.
84
Chapter 5
(a) Aliphatic
Figure 3. Aliphatic (a) and Aromatic (b) fractions of low API gravity crude oil.
Chromatographic Conditions
The original version of the TPHCWG analytical methodology did not include
hydrocarbons greater than carbon number 28 (C28). This is appropriate for most
refined petroleum products such as gasoline and diesel. However, the
concentration of hydrocarbons with carbon numbers greater than 28 (or even 35,
the upper limit for the current TPHCWG analytical method) can be as high as 50
to 60% in some crude oils with low API. Therefore, to conduct a better risk-
85
Chapter 5
based analysis of sites where crude oil is present, it may be necessary to be able
to detect hydrocarbons with carbon numbers greater than C35. This can be done
by modifying the gas chromatographic technique to quantify hydrocarbons up to
C44. (C4 4 is the upper limit for most gas chromatographs, and C44 is therefore
the technical limit of the instrument. Hydrocarbons >C44 may be present in
crude oils as described in Chapter 4.) The chromatographic conditions necessary
to do this are not routine in most environmental laboratories.
Because the TPHCWG analytical method is chromatography based, it is
imperative that the chromatographic system not introduce large amounts of bias.
In this analysis, the most common source of bias was found to be mass
discrimination [3]. Mass discrimination is created when the chromatographic
system, be it the chromatographic column or the detector, responds differently to
different materials based on their relative mass. In the TPHCWG method, the
use of a flame ionization detector (FID) ensures that discrimination does not
occur at the detector end of the system. In general, the response of an FID is
essentially the same for all hydrocarbons (on a weight basis). Saturated
hydrocarbons and compounds containing heteroatoms, such as benzothiophene,
have a slightly lower response than unsaturated, aromatic hydrocarbons because
of the lower carbon to hydrogen and carbon to heteroatom ratio of these
compounds.
Another source of mass discrimination can be created by injection port
conditions if those conditions are preferential to a particular class or classes or
compounds. In this case, the use of a narrow bore column [0.25 m innerdiameter (ID)] creates an injection port environment preferential to smaller,
lower boiling point compounds. Larger, heavier, higher boiling point
compounds may not enter the chromatographic column at the same rate and may
be underestimated. The use of a wider bore column (0.32 m ID) and proper
positioning of the column in the injection port can alleviate much of the mass
discrimination. Mass discrimination of less than 20% between n-C20 and n-C40
can be easily maintained. When analyzing crude oils these performance criteria
should replace that in the TPHCWG method comparing n-C20 to n-C28, as this
range is more applicable to the analysis of crude oils. Special attention and
maintenance must also be applied to the injection port.
The use of a wider bore column (e.g., 0.32 m ID) can alleviate many mass
discrimination problems, but can present new problems as well. The use of a
wider bore column will reduce the separation of early eluting compounds,
require a reduced temperature ramp rate, and result in longer analysis times. The
recommended GC conditions for crude oil analysis are as follows:
86
Chapter 5
Column:
Helium
5 to 7 mL/Min.
360 mL/Min.
35 mL/Min.
30 mL/Min.
325C
275C
Splitless Mode
4 mm ID Packed With Glass Wool
12.0 psi @ 50C
Approximately 50 cm/sec
Initial Temperature 10C
Hold for 5 Min.
Ramp @ 4C/Min. to 320C
Hold for 10 Min.
DB-5 30 m x -0.32 m ID, 0.25 m Film
Thickness
Summary
Analytical methods for determining the petroleum hydrocarbon content of soils
should be performance based, with common data quality objectives for elements
such as mass discrimination, mass balance, extraction efficiency, fractionation
efficiency, and calibration. When sites have been impacted by crude oils, oil
fractionation must be carefully determined and monitored, or risk-based
screening levels may be biased.
Benzene and toluene concentrations should be determined by a method specific
to volatile organics, such as USEPA Method 8260 [1]. These results can then be
combined with the TPH determinations for risk analysis.
87
Chapter 5
References
1.
2.
3.
88
Chapter 6
Introduction
Effective risk-based decision-making for chemicals that have been released to
the environment requires a basic understanding of the fate and transport of the
chemicals once released and how they interact with the different environmental
media (such as soil, water, air, and biota) that they contact. This information is
necessary in order to determine how people, or other receptors, might be
exposed to the chemicals and what the associated risks might be.
Exploration and production (E&P) sites are usually concerned with spills of
crude oils, condensates, or wastes that are associated with E&P activities.
Accordingly, spills of individual chemicals are rarely of concern and instead,
spills of hydrocarbon mixtures predominate. This is important to the evaluation
of the fate and transport of chemicals at E&P sites because the presence of the
hydrocarbon mixture will affect the solubility of any individual chemical and its
movement in the environment.
When a chemical is introduced to soil, it can volatilize, sorb to soil particles,
dissolve into the groundwater (or soil pore water), or remain in residual form.
This partitioning of the chemical into other media is described by a set of simple
mathematical equations. Movement of a chemical through different media
depends on the unique physical and chemical properties of the individual
chemical, the amount of the chemical in relation to the amount of total
petroleum hydrocarbons (TPH) in the spill, and the properties of the media. In
risk-based decision-making, this mo vement through the environment is typically
modeled with a variety of fate and transport models. Some basic equations that
can be used to estimate chemical transport will be presented in this chapter.
Many of the fate and transport models typically used for risk evaluation are
based on these equations.
The goal of this chapter is to provide an overview of the important fate and
transport processes that affect petroleum spills in soil, and the estimation of risk
from spills. The concept of partitioning when the chemical is part of a mixture
89
Chapter 6
Hazard Identification
Exposure Assessment
Toxicity (Dose-Response) Assessment
Risk Characterization
Chemical fate and transport processes are integral to completing the exposure
assessment. Figure 1 illustrates a simple conceptual site model (CSM) in which
potential exposure pathways are shown for an oil production site. Workers may
be exposed to surface soil, or to vapors from subsurface soil or groundwater,
while off-site residents may be exposed to dissolved chemicals in their drinking
water. Identifying exposure pathways is a critical component of risk-based
decision-making and the CSM is a very useful tool for visualizing the
exposure pathways that may exist at a particular site.
People
Exposure
Exposure
Impacted
shallow soil
Drinking
water
well
Groundwater
Hazard
Figure 1. Example of a conceptual site model at an oil production site.
90
Chapter 6
Ingestion of Soil
Dermal Contact
Inhalation of Volatile Emissions in Outdoor Air
Inhalation of Dust Emissions in Outdoor Air
The first two exposure pathways are direct pathways and no modeling is
required to estimate the concentrations of chemicals of potential concern that
receptors may contact. The latter two pathways are indirect pathways in which
the chemicals of potential concern originate in the impacted soil, but then are
transferred to air as they volatilize. The concentrations of the chemicals of
concern transferred from soil to breathing zone air must be estimated using
mathematical fate and transport models.
Subsurface Soil
Subsurface soil is usually defined as the region directly beneath the surficial soil
extending to just above the water table. A receptor is not ilkely to directly
contact subsurface soil. However, soil in this unsaturated zone may act as a
source of contamination to the ground surface via volatilization or to
groundwater via leaching.
91
Chapter 6
All three pathways are indirect and, therefore, require the use of fate and
transport models to estimate the concentrations of chemicals of potential
concern to which receptors may be exposed.
Groundwater
There are usually three groundwater exposure pathways:
Ingestion of Groundwater
Inhalation of Volatile Emissions in Outdoor Air
Inhalation of Volatile Emissions in Indoor Air
92
Chapter 6
order to model the fate of the chemical in the environment. This is important
because the chemical in the dissolved phase will move with the flow of
groundwater (if in an aquifer) or move downwards with infiltration (if in the
unsaturated zone). Similarly, only the chemicals in the vapor phase are assumed
to be able to readily diffuse and flow with the soil vapor.
Partitioning describes the relative concentration of a chemical in two or more
phases in contact with each other at equilibrium. The total mass of a chemical in
soil is the sum of the mass in the vapor phase, dissolved phase, sorbed to soil,
and any chemical in residual form (in the organic phase):
M T = M s + M w + M v + M org
(1)
where:
MT
Ms
Mw
Mv
Morg
=
=
=
=
=
Equation (1) would apply to the unsaturated zone where soil gas might be
present. If the mass of the chemical is in groundwater only, the mass would be
partitioned between the sorbed (sometimes called the solid phase), dissolved,
and organic phases:
M T = M s + M w + M org
(2)
where the parameters are as defined above. In Equations (1) and (2), Morg can
represent the mass of the pure form of the chemical, which occurs when a single
chemical is spilled, or it can represent the mass of the chemical found in a
separate-phase mixture, such as benzene in gasoline. The term, separate-phase
mixture is often called residual by professionals in the fields of risk
assessment and fate and transport. Used in this way, the term does not imply
mobility, but instead merely indicates the presence of a free product phase in the
soil. NAPL is also considered to b e residual product.
Equilibrium Partitioning Equation
Equations (1) and (2) define the components of the total chemical mass in soil,
however these equations dont show the relationships between the various
phases. Every chemical has a specific affinity for the various phases that can be
predicted with measured chemical properties and site-specific parameters. A
93
Chapter 6
chemical with a higher affinity for water will have a higher solubility than a
chemical with a lesser affinity for the water phase. Similarly, volatile
compounds have a high affinity for the vapor phase. In order to predict the
movement of the chemical in soil, the concentration in each phase must be
known. In the unsaturated zone, the concentration of a chemical in the soil can
be described as follows:
CT = ?bC s + ? wC w + ?a C v + ? orgCorg
(3)
where:
CT
Cs
Cw
Cv
Corg
=
=
=
=
=
a
w
org
single chemical was spilled) or the NAPL mixture (cm3 residual/cm3 dry
soil)
= Soil bulk density (g dry soil/cm3 dry soil)
(4)
Equation (4), often called the equilibrium partitioning equation, states that the
total concentration in soil can be partitioned among the sorbed, dissolved, and
vapor phases only; there is no residual present in the soil.
Many fate and transport models assume that the total concentration in soil can
be partitioned into three phases using Equation (4) [5]. In other words, the
models do not account for the separate oil phase. If the scenario being modeled
contains a NAPL phase, then Equation (4) will overestimate the concentrations
in each of the other three phases. Overestimation of the dissolved phase and soil
vapor phase will result in overestimation of the rate of leaching and
94
Chapter 6
Cv
Cw
(5)
where:
KH = Henrys Law coefficient for chemical [(mg/L vapor)/(mg/L water)]
Cw = Concentration of chemical in water (mg chemical/cm3 water)
Cv = Concentration of chemical in vapor (mg chemical/cm3 vapor)
KH has been measured for many chemicals and can be found in reference texts
[6,7]. It can also be estimated from the chemicals vapor pressure [8]. Higher
Henrys Law coefficients indicate chemicals with higher volatility. Values
range from 0.23 to 0.33 for benzene, toluene, ethylbenzene and xylenes (BTEX)
[7]. USEPA Region IX designates chemicals as being volatile if they have a
molecular weight less than 200 g/mole (mol) and a Henrys Law coefficient of
greater than 4.2E-4 (0.00042 units in dimensionless form) [9]. Table 1 shows
Henrys Law coefficients and other important fate and transport parameters for
BTEX.
Table 1. Chemical properties of BTEX [6,7].
Chemical
Parameters
BTEX:
Benzene
Ethylbenzene
Toluene
Xylenes
Molecular Vapor
Weight
Pressure
(g/mol)
(mmHg)
Henrys Law
Solubility Coefficient Koc
(mg/L)
(mL/g)
78.11
106.17
92.14
106.17
1770
169
530
198
95
9.6
28.2
8.06
0.227
0.328
0.276
0.293
6.6E+01
2.0E+02
1.4E+02
2.4E+02
95
Chapter 6
g
mg 1000 L 1000 g
C vapor 3 = K H C dissolved
m
L m 3 mg
(6)
106 g 10 3 L
3
g m
(7)
where:
Cppmv = Concentration of chemical in vapor reported as ppmv
MW = Molecular weight of chemical (g/mol)
24.45 = Molar volume at 25C and 1 atm (liter/mol)
RT
molar volume =
to calculate at other temperatures
P
R = Universal gas constant = 8.21E-2 [(atm liter)/(mol Kelvin (K))]
T = Temperature of the gas (K) (K = C + 273)
P = Pressure of the gas (atm)
Soil-Water Sorption Coefficient
The soil-water sorption coefficient, Kd, describes the relationship between the
concentration of the chemical sorbed to the soil to the concentration of the
chemical dissolved in water. The basic Freundlich equation is [6]:
Kd =
Cs
C wn
(8)
where:
Kd = Chemical-specific soil-water sorption coefficient [(mg chemical/g dry
soil)/(mg chemical/cm3 water)]
Cw = Concentration of chemical in water (mg chemical/cm3 water)
Cs = Concentration of chemical sorbed to soil (mg chemical/g dry soil)
n = Freundlich exponent (dimensionless)
96
Chapter 6
(9)
where:
Foc = Fraction organic carbon in soil (g organic carbon/g soil)
Koc = Chemical-specific organic carbon partition coefficient (mL/g)
The organic carbon partition coefficient can be found in many reference texts as
well as USEPA documents for specific chemicals [7,8]. The fraction of organic
carbon in soil is usually a site-specific value measured from a soil sample.
Equation (9) assumes that organic carbon is the sole predictor of chemical
sorption in soil. Clay content, the presence of charged particles and the fact that
organic carbon is somewhat difficult to measure, can cause the amount of
chemical sorbed to soil to be underpredicted, and hence the amount in the other
phases will be overpredicted. Lyman et al. [8] provides a discussion of the
increased sorption expected due to mineral content in the soil and provides some
equations for estimating the increased sorption. In some cases, it might be
important to have a better estimate of Kd than that provided by Equation (9). If
desired, Kd can be measured in the laboratory using soil gathered from the site.
Cw =
? b CT
? b K oc Foc + ? w + ? a K H
(10)
97
Chapter 6
Cv = Cw K H
(11)
where the variables are as defined previously. Note, that the units of CT and Cw
are not the usual units reported in laboratory data. The conversions are as
follows:
mg
g
?b 3 CT
3
cm
mg
k g k g 1000 cm
Cw
=
L
L ?b K oc Foc + ? w + ? a K H 1000 g
(12)
In Equation (12), the soil concentration, CT, can be taken from the unsaturated
zone or below the water table. Equation (12) can be useful to predict the
concentration of a chemical in groundwater from a measured soil concentration,
however as written, it is not applicable for cases containing residual. If NAPL is
present (and it can be difficult to detect visually), Equation (12) will
overestimate the dissolved phase concentration. The presence of NAPL can be
determined by considering the chemicals solubility in water. If the calculated
dissolved phase concentration exceeds the chemicals solubility, then equation
(12) is not applicable. Since Equation (12) assumes that the chemical can be
partitioned into three phases (solid, dissolved, and vapor), then there must be a
fourth phase present and Equation (3) should be used. It is very important to
make sure that the fate and transport model includes the residual term (and
thereby limits the dissolved phase and vapor phase concentrations) if it is likely
that NAPL may be present in the soil.
Up until this point, the equations presented have been describing the behavior of
a chemical partitioning in soil as if it were the only chemical present. In
hydrocarbon mixtures, that is not the case and the chemicals solubility will be
limited even further due to the presence of the overall TPH mixture. Raoults
Law predicts the effective solubility for chemicals that are in mixtures and
accordingly, it should be used if the chemical being modeled is part of a
mixture. The next section defines Raoults Law.
Effect of Raoults Law on Source Concentrations
In fate and transport screening models, the partition equations described earlier
are usually applied in the source region to estimate the concentration of the
chemical that is partitioning from the source to the vapor and/or dissolved
phases. Fate and transport models typically consider one chemical at a time, and
the partitioning of the chemical between water, soil vapor, and soil organic
matter is calculated individually for each chemical. This is not an accurate
approach for mixtures of petroleum hydrocarbons. When a chemical is part of a
mixture, its partitioning behavior can change dramatically. This is especially
98
Chapter 6
true for TPH, where each chemical usually represents a very small percentage of
the overall mixture.
The accurate prediction of the concentration of an organic chemical contained in
a mixture of many organic chemicals is essential to the assessment of potential
risks. Accordingly, fate and transport equations used to examine petroleum
hydrocarbons should include this mixture effect. In order to model organic
chemicals as part of mixtures, the fate and transport equations require a minor
modification. The required modification is to use Raoults Law to account for
the chemicals effective solubility in water.
Effect of Raoults Law on Dissolved Concentrations
The maximum dissolved concentration of a chemical in an organic mixture is a
function of its pure compound solubility and the mole fraction of the chemical in
the organic phase. Published pure compound solubility values are available for
many compounds of concern and are listed in Table 1 for BTEX. When a
chemical is part of an organic mixture, its solubility in water will be limited as
follows:
Cw max = S eff = x * S
(13)
where:
Cw max = Maximum dissolved concentration of chemical in water (mg/L)
x
= Mole fraction of chemical in organic chemical mixture (mole
chemical/mole oil)
S
= Solubility of pure chemical in water (mg/L)
S eff = Effective solubility of chemical in water (mg/L)
Equation (13) is known as Raoults Law and it is commonly used to predict the
dissolved concentration of a chemical in water exposed to a hydrocarbon
mixture [8]. For example, benzene with a pure compound solubility of 1,770
mg/L and a mole fraction of 0.001 in crude oil would have a maximum
dissolved concentration of 1.77 mg/L. The maximum dissolved concentration,
when calculated in this manner, is also called the effective solubility. The mole
fraction is calculated from:
x=
CT MWTPH
MWi CTPH
(14)
where:
CTPH = Total concentration of total petroleum hydrocarbon in soil (mg/kg)
CT
= Total concentration of individual chemical in soil (mg/kg)
MWTPH = Average molecular weight of hydrocarbon mixture (g/mol)
99
Chapter 6
MWi
For chemicals with similar molecular weights, the mass fraction can be
substituted for the mole fraction. For example, if benzene were reported in a
soil sample at 100 mg/kg and the TPH concentration was reported to be 5,000
mg/kg, then the mass fraction would be = 100/5,000, or 0.02. Assuming the
molecular weight of benzene and the TPH mixture were similar (perhaps the
TPH is a light, refined product), the maximum dissolved phase concentration of
benzene expected would be equal to 1,770 mg/L x 0.02 = 35.4 mg/L (where
benzenes solubility = 1,770 mg/L).
Raoults Law is valid for compounds that are liquids at room temperature. For
solid compounds, such as the larger polycyclic aromatic hydrocarbons (PAHs),
the subcooled liquid solubility is used in place of the pure chemical solubility
(S) in Equation (13). These values are also available in the literature [12].
To confirm that Raoults Law is appropriate for predicting the effective
solubility of a chemical that is present in oil when the oil is in contact with
water, 15 crude oils were analyzed for 43 aromatic compounds ranging from
benzene through 5- and 6-ringed PAHs. The oil was then placed in contact with
water and compounds present in the oil were allowed to partition into the water.
The concentrations of the aromatic compounds in both the oil and water phases
were determined. Figure 2 presents the data for all analytes in which the
dissolved concentrations were above the analytical reporting limit of 5 ppb.
Most of the results were at or below the dissolved concentration predicted by the
application of Raoults Law. This is evidence that Raoults Law can be used to
estimate the effective solubility of a chemical.
100
Chapter 6
1.0E+02
Measured
Raoult's
1.0E+01
1.0E+00
1.0E-01
1.0E-02
1.0E-03
1.0E-03
1.0E-02
1.0E-01
1.0E+00
1.0E+01
1.0E+02
Co
Ko
(15)
where:
Cw = Dissolved phase concentration (mg/L)
Co = Concentration of the individual chemical in the TPH (mg chemical/kg
TPH)
Ko = Oil-water partition coefficient (L/kg)
x = Mole fraction of chemical in TPH (moles/mole)
Equation (15) can be used to calculate Ko for any organic phase if S and x are
known. Since x is a mole fraction, an estimate of the average molecular weight
of the organic phase is required.
Ko =
MWi
106 mg/kg
Si MWTPH
(16)
101
Chapter 6
where:
MW i = Molecular weight of the individual chemical (g/mole)
MW TPH= Average molecular weight of the TPH (g/mole)
S
= Solubility of the individual chemical (mg/L)
Based on the distribution of compounds in the following hydrocarbon products,
typical values for MWTPH in g/mole are:
Gasoline
Condensate
Diesel/Distillate
Crude Oil
Gas Oil
100
150
180
200-250
300
x MW VP
R T
(17)
where:
Cv
VP
x
R
T
Equations (17) and (13) both indicate that the soil vapor concentrations are
limited by the presence of the TPH mixture because the mole fraction, x, will be
less than one. No matter how much NAPL is present in the soil, the vapor and
dissolved phase concentrations will not increase beyond their effective solubility
and effective saturated vapor concentration.
102
Chapter 6
The models used to estimate these transport processes and associated cleanup
levels do not address the movement of NAPL as a separate phase (e.g., the
vadose zone model SESOIL and the groundwater model AT123D [10]).
Groundwater dispersion models are used to estimate concentrations in
downgradient groundwater locations. These models are usually used in Tier 2
or Tier 3 risk analyses because the exposure point is located away from the
source. Some state regulatory programs, however, do include groundwater
transport to downgradient locations in Tier 1 [4].
The models commonly used for risk assessment tend to be screening models
because they are usually based on assumptions like uniform soil properties and
an unchanging groundwater flow field. Screening models are not applicable to
engineering design problems, such as the design of extraction wells, or for
complex hydrogeological flow regimes. Two examples of software that
integrate screening level fate and transport codes with risk assessment are BP
RISC and API DSS [11,10].
The next few sections of this chapter describe the key features usually found in
screening models.
103
Chapter 6
Air Models
If it is plausible that receptors could be exposed to soil emissions, then the
estimation of volatilization from the contaminant source will be important. If
the volatilization to the indoor air exposure pathway is being modeled, it often
drives cleanup levels because it can generate very low cleanup concentrations
due to deficiencies in the screening models. Screening models tend to
overestimate concentrations in air because they ignore many potentially
important processes impeding diffusion, they ignore biodegradation which can
be a significant loss mechanism for hydrocarbons, and the models are often
steady-state so they predict that the same exposure concentration will occur for
the entire exposure duration. For carcinogens like benzene, risk is a function of
total accumulated exposure and it is unlikely that a volatile emission exposure
would remain constant over a long exposure duration.
Diffusion
In screening models, the volatilization rate is usually estimated using the
following equation (from Ficks Law) (e.g., Farmers emission model [10,13],
groundwater emission model to indoor or outdoor air [11]):
E = Deff
(Cvs Cvf )
L
(18)
where:
E
Deff
Cvs
Cvf
L
K H ?T 2
T
(19)
where:
Dair
104
Chapter 6
RBSL (mg/kg)
6
5
4
3
2
1
0
0
0.1
0.2
0.3
0.4
0.5
Air-Filled Porosity
Figure 3. Sensitivity of subsurface soil RBSL for indoor air to changes in airfilled porosity.
When estimating concentrations in outdoor air, the vapor concentration in the
soil at ground surface, Cvf, is assumed to equal zero (or is considered to be very
small compared to the vapor concentration at the source). For the indoor air
scenario, if pressure-driven flow is ignored, the vapor phase concentration in the
soil just outside the building foundation is also assumed to equal zero.
105
Chapter 6
106
Chapter 6
Under typical conditions, the amount of infiltration through the unsaturated zone
will not equal the sites average precipitation because of evaporation, uptake by
plants, and runoff. Many modelers assume a value in the range of 10 to 20% of
the average precipitation as an infiltration rate. For example, if the average
precipitation is 20 inches per year, a reasonable assumption for infiltration, or
the amount or water that will reach the water table, is about 2 to 4 inches per
year. The infiltration rate can be limited by the vertical hydraulic conductivity
of the vadose zone soil. Consideration should also be given as to whether the
source is underneath a paved area (thereby reducing infiltration) or in an area
with artificially high recharge (such as a truck wash area). The American
Petroleum Institute (API) has compiled a large database with recharge estimates
from around the United States. A brief technical report is available from API
summarizing the estimation of infiltration and the infiltration and recharge
database [14].
The mass loading from the soil to groundwater is calculated from:
Q = Cw A I
1000 cm3
(20)
where:
Q
Cw
I
A
Besides being used in Tier 1 equations [1,2], Equation (20) is used by many
unsaturated zone models when they are linked with saturated zone models and
used to estimate concentrations downgradient [10,11].
The average downward seepage velocity of the water in the unsaturated zone is
calculated from [15]:
v=
I
?w
(21)
where:
107
Chapter 6
R =1+
(22a)
(22b)
where:
Foc = Fraction organic carbon in dry soil (g organic carbon/g soil)
Koc = Organic carbon partition coefficient (mL/g or m3/kg)
Kd = Inorganic distribution coefficient (mL/g)
b = Soil bulk density of the saturated zone (g/cm3)
w = Water-filled porosity (cm3/cm3)
Retardation does not affect the results of screening models with constant source
terms (steady-state models) that ignore degradation. It only becomes important
when degradation is considered. This does not mean that the input parameters
used in the retardation equation are unimportant because these variables are also
used in the equilibrium partitioning equation, where they are very important.
Biodegradation
Biodegradation is usually assumed to be a first-order reaction in screening
models, which means that it is assumed to be dependent only on time. In most
models biodegradation is assumed to occur in the dissolved phase only, Cw (e.g.,
the groundwater models in the API DSS and BP RISC [10,11]). The sorbed
phases and NAPL are assumed not to degrade. This means that biodegradation
only begins as the contaminant migrates away from the source.
108
Chapter 6
The biodegradation rate is probably one of the most sensitive input parameters
for petroleum hydrocarbons, because the hydrocarbons that tend to be soluble,
and therefore mobile, are also highly degradable [16,17]. Biodegradation is
discussed in more detail in Chapter 11.
References
1.
American Society for Testing and Materials (ASTM), Standard Guide for
Risk-Based Corrective Action Applied at Petroleum Release Sites, ASTM
E1739-5, November 1995.
2.
American Society for Testing and Materials (ASTM), Standard Guide for
Risk-Based Corrective Action, ASTM E2081-00, April 2000.
3.
4.
5.
Jury, W.A., Russo, D., Streile, G., and El Abd, H, Evaluation of Organic
Chemicals Residing Below the Soil Surface, Water Resources Research, 1,
13-20 (1990).
6.
7.
8.
Lyman, W.J., Reidy, P.J., and Levy, B., Mobility and Degradation of
Organic Contaminants in Subsurface Environments, C. K. Smoley, Inc,
Chelsea, Michigan (1992).
9.
109
Chapter 6
13. Johnson, P.C., and Ettinger, R.A., Heuristic Model for Predicting the
Intrusion Rate of Contaminant Vapors Into Buildings, Environmental
Science and Technology, 25, 1445-1452 (1991).
14. American Petroleum Institute, API Soil and Groundwater Research
Bulletin: Estimation of Infiltration and Recharge for Environmental Site
Assessment (1997). Available fo r download from www.api.org.
15. Guymon, G.L., Unsaturated Zone Hydrology, Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey (1994).
16. Howard, P.H., Handbook of Environmental Fate and Exposure Data For
Organic Chemicals (Volumes 1-3), Lewis Publishers, Chelsea, Michigan
(1989).
17. Howard, P.H., et al., Handbook of Environmental Degradation Rates, Lewis
Publishers Inc., Chelsea, Michigan (1991).
110
Chapter 7
Introduction
Over the last several years , a number of methods have been derived for
evaluating potential human health risks associated with exposures to petroleum
hydrocarbon mixtures. As described in previous chapters, the Total Petroleum
Hydrocarbon Criteria Working Group (TPHCWG) has conducted much of this
work, focusing primarily on refined petroleum products, such as gasoline and
diesel, which typically contain hydrocarbons below carbon number 35 (C35).
This chapter expands on the TPHCWGs work by taking into consideration
potential risks associated with exposures to crude oils and exploration and
production wastes which may contain heavy total petroleum hydrocarbon (TPH)
fractions (C35). The United States Environmental Protection Agency (USEPA)
and TPHCWG methodologies for developing reference doses (RfDs) and
reference concentrations (RfCs) are described. RfDs are recommended for
petroleum fractions C35 based on available toxicity studies.
111
Chapter 7
The RfD for any given chemical is an estimate (with uncertainty spanning
perhaps an order of magnitude) of a daily exposure to the human population
(including sensitive subgroups) that is likely to be without appreciable risk of
deleterious effects during a lifetime [7]. Similar to the RfD, the RfC is also
used to assess non-cancer effects. The RfC is an estimate (with uncertainty
spanning perhaps an order of magnitude) of continuous inhalation exposure to
the human population (including sensitive subgroups) that is likely to be without
appreciable risk of deleterious effects during a lifetime [7]. Usually, exposures
that are less than the RfD or RfC are not likely to be associated with health risks
[2,8].
Derivation of RfDs and RfCs
Oral and dermal RfDs are calculated using the following equation:
RfD = NOAEL or LOAEL / (UF1 x UF2 x MF)
RfDs are typically expressed in units of milligram per kilogram per day
(mg/kg/day).
The RfD is derived from a critical dose or concentration level usually described
as a no-observed-adverse-effect level (NOAEL) or a lowest-observed-adverseeffect level (LOAEL) identified from toxicity studies in scientific literature. A
suitable uncertainty factor (UF) is applied to the NOAEL or LOAEL to reflect
differences between the study conditions and the human exposure to which the
reference value will be applied. In addition, a modifying factor (MF) may be
applied which reflects the completeness of the supporting scientific database.
Inhalation RfCs are derived from essentially the same equation as that used to
develop RfDs. However, the NOAEL is converted into units of milligram per
cubic meter (mg/m3) based on the exposure level. Thus, the units for inhalation
RfCs are expressed in mg/m3.
Selection of Appropriate Data for Development of RfDs and RfCs
The first step to develop an RfD or an RfC is to identify a critical study and
determine the NOAEL. The NOAEL is the highest dose at which no adverse
effects are observed. If the NOAEL is not available, then a LOAEL can be
used. However, use of the LOAEL adds an additional uncertainty factor into
development of the RfD or RfC. NOAELs and LOAELs are usually based on
laboratory dose-response experiments on animals that are exposed to relatively
high doses. To develop an oral RfD the most appropriate source of the NOAEL
or LOAEL is a chronic oral study. If there are no chronic oral data available,
then subchronic oral data can be used instead. Again this adds another level of
uncertainty into the derivation of the RfD. In some cases, chronic and
subchronic inhalation studies can be used to develop oral RfDs when no oral
112
Chapter 7
data are available. Most other toxicity data (i.e., dermal, acute, or genotoxic) are
not recommended for use in the development of RfDs. Oral studies are not used
in the development of inhalation RfCs.
Uncertainty and Modifying Factors
In general, UFs are applied in multiples of 10 and can range from 1 to 10,000.
Descriptions of some UFs are as follows:
Use an additional 10-fold factor when deriving the RfD or RfC from a
LOAEL instead of a NOAEL. This factor is intended to account for the
uncertainty involved in extrapolating from LOAELs to NOAELs and is
referenced by USEPA as l0L.
113
Chapter 7
114
Chapter 7
Aliphatic Aliphatic
Oral RfD Inhalation
5.0
18.4
Neurotoxicity
C>8-C10 0.04
0.2
Decreased
Body Weight
0.1
1.0
Hepatic and
Hematological
Changes
NA
Nephrotoxicity 2.0
NA
Hepatic
(Foreign Body
Reaction)
Granuloma
C>10-C12
C>12-C16
C>16-C21 0.03 2
C>21-C35
1) Carbon range: equivalent carbon range [12].
2) This is the pyrene (C16) value.
As shown in Table 1, the TPHCWG developed oral RfDs for aliphatic and
aromatic petroleum fractions ranging from C5-C35 and inhalation RfCs for the
aliphatic and aromatic fractions from C5-C16 [13]. Due to the limitations of the
experimental design of available toxicity studies, the TPHCWG did not develop
dermal RfDs for the petroleum fractions in these carbon ranges.
Based on both observed toxicity and uncertainty associated with the available
studies, the RfDs for the aromatic fractions are at least an order of magnitude
lower than those for the aliphatic fractions. Although more toxicity data were
available for the aliphatic fractions, the data available for aromatic fractions
indicate greater toxicity. In addition, greater uncertainty was factored into the
development of the aromatic fraction RfDs than into the aliphatic fraction RfDs
because toxicity data were limited on the aromatic fractions from C9-C16 and
from C17 -C35.
For the C17-C35 aromatic fraction, the TPHCWG assigned an oral RfD of
0.03 mg/kg-day based on the indicator compound pyrene with a carbon number
of 16 (C16). This was necessary because no other toxicity data were available
for any aromatic compound with a higher carbon number, with the exception of
carcinogenicity data on polycyclic aromatic hydrocarbons (PAHs). The
aromatic fraction C17-C35 contains PAHs that are normally considered to be
indicator compounds for risk assessment purposes .
115
Chapter 7
116
Chapter 7
117
Chapter 7
Summary
Literature searches for toxicity data on heavy petroleum hydrocarbons (i.e.,
C35) were conducted and USEPA approved methods for developing RfDs were
followed. This work followed the same general procedures used previously by
the TPHCWG for developing petroleum fraction-specific RfDs. RfDs were
developed for separate aromatic and aliphatic fractions containing hydrocarbons
from C35 -C44. For the >C44 fraction, one RfD was developed for the combined
aliphatic and aromatic fraction, known as the vacuum residuum fraction.
For the aliphatic hydrocarbon fraction C35-C44, an oral RfD of 20 mg/kg/day was
developed from studies of high molecular weight white oils and microcrystalline
waxes. For the aromatic hydrocarbon fraction C35-C44, there were no data
available that could be used to develop either a dermal or an oral RfD. Lacking
any data for the aromatic hydrocarbon fraction C35-C44, an RfD of 0.03
mg/kg/day could be assigned, based on the RfD for pyrene, a C16 component of
petroleum. The TPHCWG used this approach for the C17-C35 aromatic fraction.
Based on negligible mobility, there is no scientific advantage to separating
aromatic from aliphatic hydrocarbon fractions above C44. A preliminary dermal
RfD of 0.8 mg/kg/day can be derived for a single hydrocarbon fraction (>C44).
This dermal RfD is based on dermal studies on vacuum residuum, which
contains aliphatic and aromatic compounds as well as resins and asphaltenes. It
is believed that using vacuum residuum as a surrogate for this fraction of crude
oil is conservative.
There were no oral toxicity studies found that could serve as the basis for an oral
RfD. It might be possible to develop an oral RfD based on the dermal data from
the vacuum residuum studies. However, the USEPA has not traditionally
accepted oral RfDs based on dermal data. If an oral RfD is developed based on
the dermal data from the vacuum residuum studies, then it may be reasonable to
assume an UF of 10 for dermal-to-oral bioavailability. This would suggest that
an oral RfD of 0.08 mg/kg/day could be developed for the single petroleum
fraction >C44. However, since this approach is not recommended by the
USEPA, the Petroleum Environmental Research Forum (PERF) 97-08 project
team chose to use a very conservative approach by applying an oral RfD of
0.03 mg/kg/day (based on the oral RfD of pyrene) and a dermal RfD of
0.8 mg/kg/day for the >C44 fraction.
118
Chapter 7
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
119
Chapter 7
120
Chapter 8
Introduction
A review of upstream regulations summarized in Chapter 2 indicated that there
is a need to develop a risk-based approach for establishing total petroleum
hydrocarbon (TPH) regulatory limits, particularly for crude oils, condensates,
and their associated wastes at exploration and production (E&P) sites. To
accomplish this goal, a technically sound approach for determining the potential
risk of crude oils and condensates based on their composition must first be
developed, and then this risk-based approach to assessment and management
must be accepted by upstream environmental regulators. TPH risk-based
screening levels (RBSLs) address the risk of adverse non-cancer human health
effects associated with the bulk of the hydrocarbons in complex petroleum
mixtures. TPH RBSLs are first determined for each TPH equivalent carbon
fraction and then a single TPH RBSL is calculated based on the specific mixture
or oil composition, taking into account all of the individual TPH fractions. This
chapter will describe the TPH RBSLs developed for use in the United States
based on the composition of 70 crude oils and 14 condensates.
121
Chapter 8
Impacted
soil
Volatilization
Leaching
Dissolution
Groundwater
122
Chapter 8
values used to estimate exposure are presented in Tables 1 and 2. These are also
generally consistent with those presented in the ASTM standard; however, a few
modifications have been made to better reflect current risk assessment practices
in the United States . Recent scientific data and regulatory guidance suggest that
revision of some of the default values originally presented in the ASTM
standard is appropriate. Two variables, the relative dermal absorption factor
(RAFd) and the skin adherence factor (M), were updated to better reflect more
current state and United States Environmental Protection Agency (USEPA)
default values [2,3,4,5,6,7,8,9,10]. In the absence of chemical-specific
information, an RAFd of 0.1 and an M of 0.2 mg/cm2 were assumed for all
organic chemicals. These values are consistent with current United States
federal and state guidance.
Table 1. Summary of dermal absorption factors and skin adherence values used
by selected regulatory agencies.
Chemical
Organics
Polyaromatic Hydrocarbons
All Chemicals
123
Chapter 8
Definition
Ingestion Rate, Soil
Inhalation Rate (Outdoor)
Soil to Skin Adherence
Skin Surface Area
Body Weight
Exposure Time
Exposure Frequency
Exposure Duration
Averaging Time (Non-carcinogens)
Units
mg/day
m 3/day
mg/cm2
cm2
kg
hours/day
days/yr
years
years
Residential
Symbol Value
IRsoil
191
IRsoil
8.3
M
0.2
SA
2800
BW
15
ET
24
EF
350
ED
6
AT
6
NonResidential
Value
50
31
0.2
3300
70
8
250
25
25
124
Chapter 8
125
Chapter 8
Table 3. TPH risk-based screening levels for five refined petroleum products.
Gasoline
Diesel
Baby Oil
Mineral Oil
Vaseline
Leaching to
GW (mg/kg)
RES*
RES
RES
RES
RES
Non-Residential Scenario
Outdoor
Surficial
Vaporization (mg/kg) Soils (mg/kg)
RES
30,000
RES
42,000
RES
269,000
RES
313,000
RES
333,000
RES or residual saturation indicates that the oil does not pose a significant risk
to human health even when present as residual material in soil.
Non-residential TPH RBSLs for each of the 70 crude oils analyzed as part of
this study are plotted versus their API gravity in Figure 2. These RBSLs are
based on direct human contact with contaminated surface soils. Leaching to
groundwater and volatilization to outdoor air are not exposure pathways of
concern for crude oils because most of the equivalent carbon fractions found in
crude oils are either not soluble or volatile enough to cause a concern.
80,000
70,000
60,000
50,000
40,000
30,000
20,000
10,000
0
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
API Gravity
Figure 2. Non-residential surface soil TPH RBSLs calculated for 62 crude oils
from around the world plotted by their API gravity. (API gravity data are not
available for 8 of the 70 crude oils analyzed.)
126
Chapter 8
The non-residential surficial soil RBSLs are greater than or equal to 35,000 mgTPH/kg-soil as shown in Figure 2. The highest RBSLs were 67,300 and 67,900
mg/kg, which were obtained for two very waxy crude oils containing high
amounts of the larger molecular weight aliphatic hydrocarbons (these
hydrocarbons are the least toxic). While there does not seem to be a good
correlation between calculated RBSLs and API gravity, the data are plotted in
this manner to illustrate the range of RBSLs obtained and the diversity of crude
oil types analyzed.
RBSLs for Crude Oil Associated Wastes
Wastes from four different production areas plus oily soil or waste samples were
analyzed for their equivalent carbon fractions. The analytical results were then
used to calculate TPH RBSLs for each waste and the results are shown in
Table 4. The non-residential RBSLs for the associated wastes are compared to
those of the crude oils produced at the four sites.
Table 4. Non-residential RBSLs for crude oils and their associated wastes.
Leaching to
GW (mg/kg)
Outdoor
Vaporization
(mg/kg)
Non-Residential
Surficial Soils
(mg/kg)
RES*
RES
RES
RES
RES
RES
RES
RES
67,000
89,000
80,000
79,000
RES
RES
46,000
RES
RES
RES
RES
50,000
53,000
Field #3
Crude Oil
Oily Soil
RES
RES
RES
RES
48,000
59,000
Field #4
Crude Oil
Oily Soil
RES
RES
RES
RES
50,000
58,000
Field #1
Crude Oil
Tank Bottoms
Oily Soil
Oiled Road Material
Field #2
Crude Oil
Cyclone Separator
Sludge
Slop Oil
127
Chapter 8
Table 4 (Continued).
associated wastes.
Field #5
Tank Bottoms
Composted Tank
Bottoms
Leaching to
GW (mg/kg)
Outdoor
Vaporization
(mg/kg)
Non-Residential
Surficial Soils
(mg/kg)
RES
RES
RES
RES
56,000
52,000
RES or residual saturation indicates that the oil does not pose a significant risk
to human health even when present as residual material in soil.
For Field #1, the RBSLs for tank bottoms, soil from an old spill site (oily soil),
and oiled road material are shown in comparison to the crude oil RBSL. This oil
has an API gravity of 32.3 and is a waxy/paraffinic oil. The RBSLs for the
wastes are higher than the crude oil RBSL. For Field #2, the RBSLs for cyclone
separator sludge and slop oil are shown in comparison to that of the crude oil.
This crude oil is a heavy oil, which has an API gravity of 19.4. The RBSLs for
wastes at both Field #1 and #2 are higher than that of the parent crude oils,
largely due to the loss of volatile aromatic hydrocarbons from the wastes due to
natural weathering processes . Weathering reduces the amounts of volatiles and
increases the relative amount of waxes (high molecular weight aliphatics).
Oil impacted soil was collected at two different producing oilfields (Field #3,4)
in the United States . These samples were analyzed, along with crude oil from
each field, for their equivalent carbon number fractions. Table 4 lists the nonresidential RBSLs for each of these samples. The oily soil samples yield
consistently higher RBSLs compared to the parent crude oil from the field. As
with the wastes from Fields #1 and #2, the higher RBSLs for the oily soils are
due to the loss of volatile aromatic hydrocarbons from the soil. There is a
definite trend for E&P wastes (tank bottoms, oiled road material, and oily soils)
to have higher RBSLs than their parent crude oils. Therefore, the crude oil
RBSLs presented in this chapter should be conservative for E&P wastes
accidentally released or intentionally applied to soil (for example, landfarming).
RBSLs for Biodegraded Oily Wastes
The data from Field #5 (see Table 4) are for tank bottoms from a crude oil
storage tank, and the same material after composting to reduce the hydrocarbon
content via biodegradation. After composting, the RBSL was slightly lower,
52,000 mg/kg rather than 56,000 mg/kg TPH for the tank bottoms . This is
because the more recalcitrant hydrocarbons (the large molecular weight
aromatics) are present at higher concentrations as an overall percentage of the
128
Chapter 8
oil, and this tends to drive the RBSL lower. The tank bottoms initially contained
498,000 mg/kg TPH and after composting the TPH level was reduced to 11,000
mg/kg, so the RBSL could be easily met via biodegradation processes/
composting at this site.
RBSLs for Condensates
TPH RBSLs were developed for the 14 gas condensate samples. Figure 3
illustrates the results for non-residential surface soil exposure.
TPH Non-residential Surface Soil RBSL, mg/kg soil
140,000
120,000
100,000
80,000
60,000
40,000
20,000
0
0
10
20
30
40
50
60
70
80
API Gravity
Chapter 8
particles, and dermal contact). The 95th percentile values for the crude oil
RBSLs are also shown in Table 5. These 95th percentile values are
recommended for establishing generic TPH Tier 1 RBSLs for all crude oil types,
rather than the mean value or the minimum value. At individual E&P sites, use
of generic TPH measurements should be sufficient to demonstrate whether or
not Tier 1 levels have been attained. Since these 70 crude oils are representative
of the composition of all crude oils produced around the world, there is no need
to routinely analyze E&P site soils by the detailed fractionation analytical
technique.
Table 5. Residential and non-residential RBSLs for crude oils and condensates
(all data in mg/kg soil).
Residential Summary
Statistics (All Pathways)
Crudes
Condensates
4,200
7,200
2,800
5,000
Mean
95%
Range
Minimum 2,700
Maximum 6,300
5,000
18,000
Non-Residential Summary
Statistics (All Pathways)
Crudes
Condensates
48,300
52,500
41,300
39,600
35,000
67,300
39,100
117,000
Conclusions
TPH RBSLs have been calculated for crude oils based upon the specific
composition of each oil. Since the crude oils analyzed to date reflect the
composition of all crude oils produced around the world, there is no need to
routinely analyze E&P site soils by a detailed fractionation analytical technique.
Also, the RBSLs for E&P wastes were shown to be higher than the RBSL for
fresh crude oil within a given production field. Therefore, the TPH RBSLs
developed for crude oils are conservative for E&P associated wastes. The 95%
RBSLs are recommended for use at crude oil and gas condensate spill sites.
These RBSLs can be compared to TPH results obtained using simple,
inexpensive analytical methods that have been used historically at E&P sites
(these are described in Chapter 4 and include USEPA 418.1 and 8015).
There is no reason to perform the costly TPHCWG analytical fractionation
method at most E&P sites, unless there is a need to do a site specific analysis.
For example, site specific TPH fractionation data would be desirable if there is
reason to believe that the Tier 1 RBSLs are too conservative, or if the spill
occurs on residential property. This suggests that an E&P site can be adequately
managed using generic TPH analyses (e.g., EPA Method 8015, Texas 1005, or
other TPH measurements). These TPH results can then be compared to the
Tier 1 RBSL values for crude oil and gas condensates presented in Table 5.
130
Chapter 8
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10. United States Environmental Protection Agency, Region VI, Human Health
Medium-Specific Screening Levels, Dallas, Texas, September 2000.
11. McMillen, S.J., Magaw, R.I., Kerr, J.M., and Edwards, D.A., SPE Paper No.
52722, Risk-Based Screening Method for Evaluating Total Petroleum
Hydrocarbons at Upstream Sites, presented at the 1999 SPE/EPA
Exploration and Production Environmental Conference, Austin, Texas (1999).
12. United States Environmental Protection Agency, Office of Emergency and
Remedial Response, Risk Assessment Guidance for Superfund Volume I:
Human Health Evaluation Manual, September 29, 1989.
13. United States Environmental Protection Agency, Office of Research and
Development, Exposure Factors Handbook, EPA/600/P-95/002Fa,
August 1997.
131
Chapter 9
Introduction
The transport of hydrocarbons to groundwater is an important remediation
consideration for many petroleum release sites. There is often a concern that
hydrocarbons may leach from impacted soil, migrate to groundwater, and impact
the water source at unacceptable levels. Simple screening level fate and
transport models that do not account for Raoults Law suggest that this potential
exposure pathway may be of concern at many crude oil release sites.
Incorporating Raoults Law into the analysis indicates that crude oils generally
have a limited potential to impact groundwater at levels that would present
unacceptable human health risks.
(1)
where:
Cw max = Maximum dissolved concentration of chemical in water (mg/L)
x
= Mole fraction of chemical in organic chemical mixture (mole
chemical/mole oil)
S
= Solubility of pure chemical in water (mg/L)
S eff = Effective solubility of chemical in water (mg/L)
Equation (1) is known as Raoults Law and it is commonly used to predict the
dissolved concentration of a chemical in water in contact with a hydrocarbon
mixture. Raoults Law is valid for compounds that are liquids at room
temperature. For solid compounds, such as the larger polycyclic aromatic
hydrocarbons (PAHs), the subcooled liquid solubility is used in place of the pure
chemical solubility (S) in Equation (1).
132
Chapter 9
When petroleum hydrocarbons (for example crude oil) are present as nonaqueous phase liquid (NAPL) in contact with groundwater, the dissolved
concentrations of the individual chemicals are generally not affected by the
amount of oil in contact with the water as illustrated below1:
Mass of crude oil = mass of water:
Crude
1,000 mg/kg Benzene
Water
1.8 mg/L Benzene
Only when the mass of the crude oil is very low does the amount of the crude
oil present affect the effective solubility of individual compounds. At low oil
concentrations, lower final concentrations in both phases can be expected due to
partitioning into the soil organic matter as illustrated in Figure 1.
133
Chapter 9
Benzene
Naphthalene
Chrysene
Benzo(a)pyrene
Co max.
(mg/kg )1
5,900
3,700
120
7.7
X max. 2
1.51E-02
5.77E-03
1.05E-02
6.11E-06
S3
(mg/L)
1.80E+03
1.03E+02
1.65E-01
4.00E-02
Cw max.
(mg/L)4
2.72E+01
5.94E-01
1.74E-05
2.4E-07
MCL 5
(mg/L)
5.0E-03
NA
NA
2.0E-04
PRG6
(mg/L)
3.5E-04
6.2E-03
9.2E-03
9.2E-06
>MCL or
PRG
Yes
Yes
No
No
134
Chapter 9
PAH
Naphthalene
Acenaphthene
Anthracene
Fluorene
Benzo[a]anthracene
Fluoranthene
Chrysene
Pyrene
Dibenz[a,h]anthracene
Benzo[a]pyrene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Indeno[1,2,3cd]pyrene
Cw
Co max.
S
max.
1 X max.2
3
(mg/L)4
(mg/kg )
(mg/L)
3,700
5.77E-03 1.03E+02 5.9E-01
MCL 5
(mg/L)
NA
PRG6
(mg/L)
6.2E-03
>MCL
or PRG
Yes
58
17
380
38
26
120
82
9.2
7.7
14
7
1.7
NA
NA
NA
NA
NA
NA
NA
NA
2.0E-04
NA
NA
NA
3.7E-01
1.8E+00
2.4E-01
9.2E-05
1.5E+00
9.2E-03
1.8E-01
9.2E-06
9.2E-06
9.2E-05
9.2E-04
9.2E-05
No
No
No
No
No
No
No
No
No
No
No
No
7.52E-05
1.91E-05
4.57E-04
3.33E-05
2.57E-05
1.05E-04
8.1E-05
6.60E-06
6.11E-06
1.11E-05
5.55E-06
1.23E-06
2.12E+01
4.43E+00
1.24E+01
2.35E-01
1.08E+00
1.65E-01
1.04E+00
6.23E-01
4.00E-02
3.85E-02
6.15E-02
5.50E-04
1.6E-03
8.5E-05
5.7E-03
8.0E-06
2.8E-05
1.7E-05
8.4E-05
4.0E-06
2.4E-07
4.2E-07
3.4E-07
6.8E-10
135
Chapter 9
GW
X=
GWIng * LDF
S
(2)
where:
GW
GW Ing 1
(mg/L)
2.4
0.98
0.98
0.98
0.73
0.73
1.5
1.5
2.4
2.4
2.4
49
S2
(mg/L)
530
65
25
5.8
0.65
0.0066
36
5.4
0.43
0.034
0.00076
2.5E-06
Concentration in
TPH (Mole Fraction)
0.09
0.30
0.78
1
1
1
0.83
1
1
1
1
1
Chapter 9
Water
1 mg/L
In unimpacted soil, both the soil organic content (SOC) and water are free of
petroleum hydrocarbons. When oil is added to soil, an individual chemical
contained in the oil will partition between the oil, SOC, and water. At low oil
levels, the concentration of the chemical in the oil phase decreases due to loss to
the SOC and water. This results in a lower dissolved concentration than would
be expected for oil-water partitioning alone. As the mass of oil increases
relative to the SOC, the final organic phase concentration remains closer to the
initial oil concentration. The dissolved concentration gets closer to, but cannot
exceed, the effective solubility predicted by Raoults Law.
The partitioning between oil, SOC, and water is demonstrated in the following
example. The concentration of the chemical of interest is initially 1,000 mg/kg
in the oil phase. Raoults Law predicts a maximum dissolved concentration of
10 mg/L. The amount of oil ranges from 100 to 10,000 mg/kg soil. The SOC is
1,000 mg/kg soil. Co f is the final concentration of the compound in the oil
phase. Cs is the concentration in the SOC phase and Cw is the dissolved phase.
137
Chapter 9
Amount
of SOC
(mg/kg)
1,000
1,000
1,000
Final Conc. in
Oil Phase (Cof
mg/kg)
118
571
930
Conc. of Chemical
in SOC Phase
(C s mg/kg)
88
429
698
Conc. of
Chemical in
Dissolved Phase
(Cw mg/L)
1.18
5.71
9.30
Figure 1 shows the same effect over a wider range of concentrations. The
dissolved concentration of the compound increases with oil concentrations at
low levels, but the maximum dissolved concentration remains limited to the
value predicted by Raoults Law.
100.00
Cw (mg/L)
10.00
Cw
Effective
Solubility
Predicted
by
Raoult's
Law
1.00
0.10
10
100
1000
10000
100000
1000000
Chapter 9
(3)
where:
Cs = Soil screening level (mg/kg)
Cw = Dissolved groundwater protection goal (mg/L)
Kd = Soil-water sorption coefficient (mg/kg dry soil)/(mg/
L water)
T w = Water filled porosity (cm3 water/cm3 soil)
T a = Air filled porosity (cm3 air/cm3 soil)
KH = Henrys Law coefficient (mg/L vapor/mg/L water)
?b = Soil bulk density (g dry soil/cm3 soil)
The component of the equation (T w + Ta KH )/?b takes into account the
fraction of the COCs in the water and vapor phase. Since this term is typically
much smaller than Kd, Equation (3) can be simplified for this discussion to:
Cs = Cw * Kd
(4)
(5)
where:
Koc = SOC / water partition coefficient (mg /kg SOC)/(mg /L water)
foc = The fraction of organic carbon in the soil (kg SOC/kg soil)
If two organic phases are present, the organic carbon in the soil and the oil, then
the soil-water sorption coefficient in Equation (4) can be redefined as:
Kd = (Koc1 * foc1) + (Koc 2 * foc2)
(6)
An oil impacted soil has two organic phases, the SOC and the oil. By replacing
Koc2 with Koil (mg/kg oil)/(mg/L water), and foc 2 with the oil concentration in
the soil (kg oil/kg soil), Kd from Equation (6) can be inserted into Equation (4):
Cs = Cw * ((Koc * foc) + (Koil * foil))
(7)
139
Chapter 9
Koil =
MWi
10 6 mg / k g
S i MWo
(8)
where:
MW i = Molecular weight of the compound (g/mole)
MW o = Average molecular weight of the TPH (g/mole)
Si
= Solubility of the compound of interest (mg/L)
Summary
The protection of groundwater is a critical component of environmental
management at sites impacted with petroleum hydrocarbons. The potential for
chemicals of concern to leach from oil or petroleum impacted soil must be
understood in order to accurately assess the risk to human health from the
ingestion of groundwater. In this chapter, the technical basis for calculating
maximum dissolved hydrocarbon concentrations in leachate was presented. If
the concentration of a constituent in either the oil or soil phase is known, the
maximum dissolved concentration can be calculated.
The dissolved
concentration is limited by equilibrium partitioning between the oil and water
phases. Partitioning between soil organic matter and these phases can further
reduce the dissolved hydrocarbon leachate concentration. The analysis of crude
oil data presented in this chapter illustrates that benzene and naphthalene may
potentially leach from crude oil-containing soil to groundwater at levels that
may pose an unacceptable risk to human health. The other priority pollutant
PAHs are not present in crude oil at high enough concentrations to be a potential
threat to groundwater via leaching from soil.
References
1.
2.
EPA Region IX, PRGs Table 2000 Update, San Francisco, California,
November 3, 2000.
3.
Kerr, J.M., Hamed, M.M., Melton, H.R., Magaw, R.I., McMillen, S.J., and
Naughton, G., Risk-Based Soil Screening Levels for Crude Oil: Role of
Polyaromatic Hydrocarbons, in the Proceedings of the 6th International
Petroleum Environmental Conference, Houston, Texas, K.L. Sublette, ed.,
SCG, Inc., Tulsa, Oklahoma, 474-488 (2000).
4.
140
Chapter 9
5.
Peters, C.A., Mukherji, S., Knightes, C.D., and Weber, W.J., Phase
Stability of Multi-Component NAPLs Containing PAHs, Environmental
Science and Technology, 31, 2540-2546 (1997).
6.
7.
8.
Texas Natural Resource Conservation Commission, Subchapter D Development of Protective Concentration Levels, in Texas Risk Reduction
Program (TRRP) Rule, 30 TAC 350 (1999).
9.
141
Chapter 10
Introduction
Polyaromatic hydrocarbons (PAHs) are ubiquitous in the environment and result
from both naturally-occurring and man-made sources. PAHs can be formed
whenever organic substances are exposed to high temperatures. Some PAHs are
also synthesized by several species of plants and bacteria [1]. The largest single
source of PAHs to the environment is likely the residential burning of wood [2].
Combustion of fossil fuels is also a major source, with other sources including
volcanoes , agricultural burning, asphalt roads, and forest fires [2]. In the home,
PAHs can be found in tobacco smoke, grilled and smoked foods, wood-burning
fireplaces, meat, processed or pickled foods, and beverages.
Most of the direct releases of PAHs are to the atmosphere, and most of these are
associated with particulate matter. Settling of the particulates allows PAHs to
be redistributed to other environmental media, such as soil and water. PAHs are
found in relatively high concentrations at certain types of industrial sites,
particularly in contaminated soils at manufactured gas plants (MGP) and woodtreatment facilities. Creosote, a common wood preservative, is a distillate of
coal tar (distilled from coal at high temperatures in the absence of oxygen), and
contains higher concentrations of PAHs than many other substances [3].
At sufficient dose levels, laboratory studies show that some PAHs cause adverse
health effects including cancer and reproductive difficulties in animals. People
exposed for long periods of time to mixtures of chemicals that include PAHs can
also develop cancer [2]. Therefore, PAHs have been identified as chemicals of
potential environmental concern, and PAHs known to cause adverse effects have
been placed on the United States Environmental Protection Agency (USEPA)
priority pollutant list.
The structures of the 16 PAHs currently on the USEPA priority pollutant list are
shown in Figure 1. Selected physical-chemical properties for these PAHs are
given in Table 1. Boiling points range from 217 to 542C and aqueous
solubilities of pure solids range from 0.00026 to 31 mg/L.
142
Chapter 10
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Anthracene
Phenanthrene
Fluoranthene
Pyrene
Benz [a]anthracene
Chrysene
Benzo [b]fluoranthene
Benzo [k]fluoranthene
Benzo [a]pyrene
Indeno [1,2,3-cd]
pyrene
Dibenz [a,h]anthracene
Benzo [g,h,i]perylene
PAH
Naphthalene
Acenaphthene
Acenaphthylene
Anthracene
Phenanthrene
Fluorene
Fluoranthene
Benz[a]anthracene
Chrysene
Pyrene
Benzo[a]pyrene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Dibenz[a,h]anthracene
Benzo[g,h,i]perylene
Indeno[1,2,3 -cd]-pyrene
1)
Boiling
Point 1 (C)
217
279
265
340
340
293
295
400
448
360
496
481
480
535
542
530
Aqueous
Solubility of Pure
Solid (mg/L)
31
3.8
16.1
0.045
1.1
1.9
0.26
0.011
0.0015
0.132
0.0038
0.0015
0.0008
0.0005
0.00026
0.062
Aqueous Solubility
of Pure Subcooled
Liquid (mg/L)
103.23
19
73.094
4.59
4.07
11.875
5.26
0.275
0.1545
1.015
0.1265
0.0384
0.0615
0.125
0.0866
1.55
Kp1
log
3.37
4.33
4.07
4.45
4.46
4.18
5.33
5.61
5.61
5.32
6.04
6.57
6.84
5.97
7.23
7.66
Vapor
Pressure
(Torr at 20C)
4.92 x 10-2
2.0 x 10-2
2.9 x 10-2
1.96 x 10-4
6.80 x 10-4
1.3 x 10-2
6.0 x 10-6
5.0 x 10-9
6.3 x 10-7
6.85 x 10-7
5.0 x 10-7
5.0 x 10-7
5.0 x 10-7
1.0 x 10-10
1.0 x 10-10
1.0 x 10-10
PAHs occur naturally in crude oils. While crude oils and their associated wastes
contain PAHs, there are few published data on the amounts of priority pollutant
143
Chapter 10
Analytical Methodology
Sixty crude oils from production sites around the world were analyzed for the 16
priority pollutant PAHs. Figure 2 illustrates the sources of the crude oils, by
geographic region. The crude oils were selected based on their diversity in API
gravities and compound classes. The goal was to analyze a diverse group of oils
that would represent the range of crude oil types produced around the world. In
addition to the 60 crude oils, 10 condensates, 10 oily wastes (tank bottoms,
sludges, etc.), and 6 oil-containing soils also were analyzed for their PAH
content.
144
Chapter 10
10
9
3 1
1
8
2
53
3
33
3
Figure 2. Number of crude oil samples (60 total) analyzed by the geographic
region from which they originated.
Using pentane, a dilution was prepared for each crude oil and condensate to
obtain an approximate concentration of 5 mg/mL. Each dilution was then
spiked directly with the following surrogates; naphthalene-d8, acenaphthene-d10,
phenanthrene-d10, and benzo[a]pyrene-d12. Extracts were obtained from wastes
and soils using either methylene chloride or pentane as the solvent. The diluted
crude oils and extracts were then passed through a silica gel chromatography
column to separate the aromatic hydrocarbons using modified USEPA Method
3630.
Pentane was used to elute the saturate fraction, methylene
chloride:pentane (50:50) was used to elute the aromatic hydrocarbons. The
aromatic fraction was concentrated and spiked with the following internal
standards; chrysene-d12 and fluorene d10.
To determine the concentrations of selected PAHs, the concentrated extracts and
diluted crude oils were analyzed by gas chromatography/mass spectrometry
(GC/MS) operated in the selected ion monitoring (SIM) mode using a modified
USEPA Method 8270. A five-point calibration, an Instrumental Reference
Material (NIST IRM 1491), and an oil reference standard (Alaska North Slope
crude oil) were analyzed at the beginning of each instrumental sequence. All
instruments were calibrated with analytical standards prior to the analysis of the
samples. Target analyte concentrations of the 16 priority pollutant PAHs were
calculated versus the internal standard compound and were corrected for
recovery efficiency of the surrogate compounds. The recovery of the surrogate
compounds was calculated relative to the internal standards added to the
samples prior to instrumental analysis.
145
Chapter 10
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Anthracene
Phenanthrene
Fluorene
Benz[a]anthracene
Fluoranthene
Chrysene
Pyrene
Dibenz[a,h]anthracene
Benzo[a]pyrene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Indeno[1,2,3-cd]pyrene
Benzo[g,h,i]perylene
Minimum
(mg/kg)
1.2
ND
ND
ND
ND
1.4
ND
ND
4
ND
ND
ND
ND
ND
ND
ND
Maximum
(mg/kg)
3700
NA
58
17
916
380
38
26
120
82
9.2
7.7
14
7
1.7
9.6
Mean
(mg/kg)
422.9
NA
13.9
3.4
176.7
73.6
5.5
3.9
28.5
15.5
1.0
2.0
3.9
0.46
0.06
1.53
Detection
Frequency
(%)
100
0
80
40
98
100
67
40
100
97
47
75
100
93
7
63
146
Chapter 10
Table 2b. PAH content of 10 condensates (all data are reported in mg/kg
condensate).
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Anthracene
Phenanthrene
Fluorene
Benz[a]anthracene
Fluoranthene
Chrysene
Pyrene
Dibenz[a,h]anthracene
Benzo[a]pyrene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Indeno[1,2,3-cd]pyrene
Benzo[g,h ,i]perylene
Minimum
(mg/kg)
200
ND
ND
ND
ND
3.9
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Maximum
(mg/kg)
5,700
9.2
12
27
250
82
0.78
11
5.5
12
NA
NA
2
NA
NA
NA
Mean
(mg/kg)
1,690
1.15
1.43
5.91
90
44.8
0.30
2.47
1.93
2.96
NA
NA
0.64
NA
NA
NA
Detection
Frequency
(%)
100
10
10
50
90
100
10
30
40
40
0
0
30
0
0
0
147
Chapter 10
Soil
Concentrations
MGP Sites
Mean [11]
(mg/kg)
--2
--225
156
379
2,174
491
317
345
260
238
92
207
2,451
---
Sludges/
Tank
Bottoms
Mean
(mg/kg)
44.00
0.29
6.51
21.09
2.22
55.53
2.31
5.42
2.98
12.16
1.74
0.28
0.97
0.20
0.65
0.73
Soil
Concentrations
E&P Sites
Mean (mg/kg)
0.183
0.017
0.733
0.455
0.214
1.429
0.283
0.869
0.393
1.385
0.199
0.061
0.119
0.024
0.094
0.176
148
Chapter 10
10000
1000
100
10
1
Coal Tar
Benz[a]anthracene
Benzo[a]pyrene
Pit sludges/tank
bottoms (10)
Chrysene
Indeno[1,2,3,-cd]pyrene
Benzo[b]fluoranthene
Dibenz[ a,h]anthracene
Condensates
(10)
Benzo[k]fluoranthene
149
Chapter 10
180
160
140
120
10
15
20
25
30
35
40
45
50
API gravity
Figure 4. Concentration of 4-6 ring PAHs versus API gravity of crude oils.
Crude oils from lacustrine sources that have been biodegraded contain the
highest concentrations of >4-ring PAHs.
Figure 4 also illustrates that API gravity is not a good predictor of the
concentration of 4- to 6-ring PAHs that may be present in crude oils. Source
rock and maturity have been reported to correlate with PAH content of crude
oils [13].
150
Chapter 10
Table 4. Tier 1 RBSLs from major oil producing states showing the concentration of each PAH for soils in residential or
industrial areas, plus those concentrations of crude oil in soil that are protective of groundwater. All units are in mg/kg.
Louisiana2
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Anthracene
Phenanthrene
Fluoranthene
Pyrene
Benz[a]anthracene
Chrysene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Benzo[a]pyrene
Indeno[1,2,3-cd]pyrene
Dibenz[a,h]anthracene
Benzo[g,h,i]perylene
Alaska 1
38-43
NA
190-210
240-270
3,900-4,300
NA
NA
1,400-1,500
5.5-6
550-620
17-20
170-200
2.4-3
50-54
5-6
NA
Industrial
44
NA
3,900
3,100
25,000
NA
3,600
2,700
3.6
400
3.6
35
0.36
3.6
0.36
NA
GW
Residential Protect
6.3
1.5
NA
NA
260
220
180
230
1,400
120
NA
NA
200
1,200
150
1,100
0.56
8.6
61
76
0.56
29
5.5
120
0.33
23
0.56
9.2
0.33
540
NA
NA
New Mexico3
Industrial
18,500
NA
27,700
19,600
157,000
14,500
22,100
16,700
21.8
2,150
21.7
21.9
2.19
NA
2.21
NA
GW
Residential Protect
1,790
0.68
NA
NA
2,900
187.95
2,150
196.12
16,900
4,499.81
1,590
270.07
2,340
1,247.59
1,760
1,301.71
9.49
7.48
940
810.27
9.45
25.68
9.52
25.68
0.952
4.74
NA
NA
0.952
3.74
NA
NA
Texas4
Industrial
360
37,000
37,000
25,000
190,000
19,000
25,000
19,000
24
2,400
24
240
2.4
24
2.4
19,000
Residential
220
3,800
3,000
2,300
18,000
1,700
2,300
1,700
5.7
560
5.7
57
0.56
5.7
0.55
1,800
GW
Protect
31
410
240
300
6,900
420
1,900
1,100
18
1,500
60
620
7.6
170
15
46,000
NA = Not available.
1) Cleanup levels based on migration to groundwater assuming over 40-inch and under 40-inch rainfall zones, respectively. Cleanup levels correspond with
Alaska AK101-103AA and AK101-103 methods. Source: ADEC, 2000 [18].
2) Soil screening standards for industrial exposures, non-industrial exposures, and protection of groundwater. Source: LDEQ RECAP, 2000 [19].
3) Industrial and residential direct contact exposures, and Tier 1 levels protective of groundwater assuming a DAF = 1. Source: NMED, 2000 [20].
4) Industrial and residential direct contact exposures and protection of groundwater assuming a 0.5-acre source. Source: TNRCC, 2001 [21].
151
Chapter 10
of Louisiana uses a target risk of 1 x 10-6 for carcinogens and a Hazard Quotient
(HQ) of 0.1 for non-carcinogens, while Texas uses 1 x 10-5 for carcinogens and
a HQ = 1.0. In any case, both of these target cancer risk values lie within the
acceptable target risk range of 1 x 10-6 to 1 x 10-4 set by USEPA for evaluating
contaminated sites under Superfund [17] and are in line with target risk levels
commonly adopted by regulatory agencies for environmental programs in many
states.
The acceptable TPH level was estimated for each oil or condensate, and the
results are shown in Table 5. For the crude oils, the estimated acceptable TPH
levels ranged from 1.4 to >100%, except for the case of naphthalene. For the
condensates, the estimated acceptable TPH levels in soil range from 28 to
>100%. Again, naphthalene is an exception, resulting in an acceptable
concentration of condensate in soil as low as 0.012%. This TPH level is
protective of groundwater resources. Acceptable TPH levels for direct
residential contact with impacted s oils would be higher.
Naphthalene may be present in some crude oils and condensates at
concentrations that exceed Tier 1 PAH levels even at low TPH levels.
Naphthalene may be of particular concern for protecting groundwater resources.
However, the acceptable oil and condensate concentrations provided in Table 5
are for fresh spills. Natural attenuation of naphthalene due to weathering
(volatilization and biodegradation) may occur quickly at some spill sites. Also,
Tier 1 RBSLs for naphthalene do not consider the impact of Raoults Law as
described in Chapter 9. For these reasons, the preferred method for assessing
152
Chapter 10
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Anthracene
Phenanthrene
Fluoranthene
Pyrene
Benz[a]anthracene
Chrysene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Benzo[a]pyrene
Indeno[1,2,3-cd]pyrene
Dibenz[a,h]anthracene
Benzo[g,h,i]perylene
Maximum Observed
Lowest Tier 1 Concentration
(mg/kg Oil)
RBSL From
Crude
Table 4
(mg/kg Soil) Oils
Condensates
0.68
3,700
5,700
410
ND
9.2
187.95
58
12
180
380
82
120
17
27
270.07
916
250
200
26
11
150
82
12
0.56
38
0.78
61
120
5.5
0.56
14
2
5.5
7
ND
0.33
7.7
ND
0.56
1.7
ND
0.33
9.2
ND
1800
9.6
ND
Acceptable
Concentration for a
Fresh Spill (% Oil or
Condensate in Soil)
Crude Oil Condensate
in Soil
in Soil
0.018
0.012
>100
>100
>100
>100
47.3
>100
>100
>100
29.5
>100
>100
>100
>100
>100
1.4
71.8
50.8
>100
4.0
28.0
78.5
>100
4.2
>100
32.9
>100
3.5
>100
>100
>100
Summary
Sixty crude oils and ten condensates were analyzed for their concentrations of
16 priority pollutant PAHs. Screening of the human health risk associated with
153
Chapter 10
soil containing crude oil or condensate showed the risk for 15 of the 16 PAHs
was not significant at TPH concentrations well above 10,000 ppm, the current
TPH soil management level in many states. Naphthalene is the only PAH that
may be present in crude oils or condensates at concentrations that may pose a
risk at oil concentrations of 1% or lower. The limiting exposure pathway for
naphthalene is leaching from soil to groundwater and protection of groundwater
resources, rather than direct residential exposure to contaminated surface soils.
Overall, this work indicates that the low levels of PAHs in crude oils are
unlikely to be a major risk management consideration at crude oil or condensate
spill sites. This constitutes compelling evidence that routine analyses for all
PAHs at E&P crude oil or condensate spill sites are unnecessary. However,
naphthalene should be analyzed as part of an USEPA 8260 analysis along with
the volatiles benzene, toluene, ethyl benzene, and xylenes.
References
1.
2.
3.
4.
Lide, D.R., CRC Handbook of Chemistry and Physics , 77th edition, CRC
Press, Inc., Boca Raton (1996-1997).
5.
MacKay D., Shiu, W.Y., and Ma, K.C., Illustrated Handbook of PhysicalChemical Properties and Environmental Fate for Organic Chemicals,
Vol. 2, Lewis Publishers, Chelsea, MI (1992).
6.
7.
8.
9.
154
Chapter 10
155
Chapter 11
Introduction
Benzene is commonly found in the environment from both human activities and
natural processes. Benzene was first discovered in 1825 and isolated from coal
tar in 1849, while today it is manufactured mostly from petroleum sources.
Benzene is used by industry to make other chemicals such as ethylbenzene for
plastics manufacture, cumene for resins, and cyclohexane for nylon and
synthetic fibers [1]. Natural sources of benzene in the environment include
volcanoes, forest fires, and crude oil seeps. Benzene occurs naturally in most
crude oils, is a byproduct of oil refining processes, and also occurs in natural gas
production condensates.
Benzene is a known human carcinogen. In workers, long-term exposure to high
concentrations of benzene in air has been shown to cause cancer of the bloodforming organs. In laboratory animals, benzene has been shown to produce
several types of cancer following oral or inhalation exposure. There are still
questions concerning both the mechanisms of benzene carcinogenesis and the
most appropriate models for developing human risk estimates. These issues are
actively being studied and debated in the scientific community. In addition to
cancer, benzene is also known to produce other adverse health effects, again
principally on the blood-forming organs, although neurological and reproductive
effects may also be of concern [1]. Most people are exposed to a small amount
of benzene every day, mainly via inhalation of vapors from commercial products
such as glues, paints, cigarette smoke, and vehicle exhaust. People may come
into contact with benzene through the inhalation, ingestion, or dermal contact
exposure pathways.
Most upstream regulatory programs (such as those in the States of Texas and
Louisiana) do not routinely require benzene analysis of exploration and
production (E&P) site soils and do not routinely set regulatory limits for
benzene in soil. Upstream regulatory agencies in California, New Mexico, and
Michigan are exceptions and do require benzene analyses for soils at E&P sites.
Regulatory limits for benzene in soil are routinely set in downstream regulatory
programs , such as those with jurisdiction over underground storage tank (UST)
sites . Most often, these are based on Tier 1 Risk Based Screening Levels
(RBSLs) developed for protecting groundwater resources. In developing RBSLs,
a number of fate and transport assumptions are typically used that are now
known to be overly conservative for benzene. For example, Tier 1 RBSLs have
156
Chapter 11
157
Chapter 11
Figure 1. Sample locations for the crude oils and condensates. Twelve of the
condensates were from the United States (excluding Alaska and Hawaii).
As shown in Table 1, the highest observed concentration of benzene in the 69
crude oils was 5900 mg/kg of oil or 0.59 wt %, and the mean concentration of
benzene in the crude oils was 1,340 mg/kg. Two crude oils contained less than
1.2 mg/kg benzene (the detection limit for the analytical method). In general,
higher API gravity crude oils and condensates tend to contain more benzene as
shown in Figure 2. The condensates contained more benzene than the crude
oils, with the maximum concentration being 35,600 mg/kg of condensate
(3.56 wt %). The mean concentration of benzene for the 14 condensates was
10,300 mg/kg. There is roughly 10 times more benzene on average in the
analyzed condensates than in the analyzed crude oils.
Table 1. Concentrations of benzene in crude oils and condensates analyzed in
this study.
Concentration of Benzene (mg/kg Oil)
# of Samples
69 Crude Oils
14 Condensates
API Gravity
Range ()
8.846.4
4570.1
Mean
1,340
10,300
Median
780
6400
Minimum
ND*
1470
Maximum
5900
35,600
Number of
Samples With
Benzene = ND
2
0
158
Chapter 11
105
104
103
102
101
Benzene Conc. in Oil Data
Regression line
100
10-1
0
10
20
30
40
50
60
70
API Gravity
Figure 2. Benzene concentrations versus API gravity for 61 crude oils and 14
condensates (API gravity data were unavailable for 8 crude oils).
159
Chapter 11
160
Chapter 11
The RBSL for benzene in soil that is protective of groundwater, RBSLs-gw, was
calculated using the following equation:
RBSL s gw = CT = K T * DAF o * C R
(1)
where:
CT = The total concentration of benzene in soil based on the groundwater
exposure pathway (g/g-soil)
CR = The acceptable concentration of benzene in groundwater at the
groundwater receptor (g/cm3-water)
CR is either a risk-based screening level for water, RBSLw, or a groundwater
regulatory standard (in the United States, the maximum contaminant level for
benzene in groundwater is 5 x 10-9 g/cm3-water; the State of New Mexico has a
groundwater standard = 1.0 x 10-8 g/cm3-water).
If appropriate, RBSLw (g/cm3-w) can be calculated for benzene, a carcinogen,
according to the following equation:
RBSL w =
TR * BW * AT c * 365 * 10 6
IR w * ED * EF * SFo
(2)
where:
Target excess individual lifetime cancer risk (10-5)
Adult body weight (70 kg)
Averaging time for carcinogens (70 years)
Oral cancer slope factor [mg/kg-day]-1 (0.029 for benzene)
Adult daily water ingestion rate (2 L/day)
Exposure frequency for residents (350 days/year)
Exposure duration for adult residents (30 years)
The total partition coefficient for benzene. It is the ratio of total soil
concentration to pore water concentration in the source zone of the
contamination (cm3-water/g-soil).
DAFo = The overall dilution attenuation factor (unitless) which is defined as:
TR
BW
ATc
SFo
IRw
EF
ED
KT
=
=
=
=
=
=
=
=
(3)
where:
DAFunsat = Cs/Cunsat = Unsaturated zone dilution attenuation factor (unitless)
161
Chapter 11
KT =
CT w + K o b + K a a + K o o
=
CS
b
(4)
where:
CT
CS
b
w
Kd
Ka
o MWi
S i MWo
(5)
where:
162
Chapter 11
o
MW i
MW o
Si
=
=
=
=
The volumetric oil content in the soil can be related to the residual TPH
concentration with the following equation:
C
o = b TPH
o
(6)
where:
CTPH = Residually trapped TPH concentration in soil (g/g-soil)
Substituting Equations (5) and (6) for Koo, Equation (4) becomes:
KT =
w + K d b + K aa +
b
b MWi CTPH
S i MWo
(7)
Thus, the overall partition coefficient, KT, will be dependent upon the level of
residual TPH in the soil. (Note that the residual TPH level is the amount in
excess of the sorbed TPH level which is nominally <100 mg/kg for a low
organic carbon content soil.)
Summary of DAFunsat
The attenuation in the unsaturated zone is due to the following factors:
1) Biodegradation of the contaminant in the region beneath the source of
contamination and the groundwater table.
2) Depletion of the concentrations in the source of contamination due to
losses associated with leaching, volatilization, and biodegradation.
A commonly used approach for modeling vadose zone transport is that of nl
et al. [6] which uses the equation of van Genuchten and Alves [7]. This
equation is also the basis for modeling unsaturated zone transport in the
computer model VADSAT [6]. To determine DAFunsat values for the RBSLs
presented here a simpler model was used. This model is based on treating the
contaminated zone and the unsaturated zone beneath the contaminated zone (see
Figure 3) as two separate, completely mixed zones. (The nl model treats the
163
Chapter 11
source zone as a completely mixed zone, but models advection and dispersion in
the region below the source zone.) This completely -mixed model approach was
used here because it is computationally easy to use (it is possible to develop a
simple algebraic expression for DAFunsat) and retains the essential parameters of
the nl et al. model. For completely -mixed conditions, the DAFunsat can be
determined from the following equation:
DAFunsat =
H b K T ,unsat
u
(8)
where:
u
+ unsat
H
=
b K T , unsat
(9)
and
u
+ S
d
b K T ,S
V +
(10)
164
Chapter 11
KT,unsat = The overall soil-water partition coefficient for the unsaturated zone
(cm3-water/g-soil)
KT,S
= The overall soil-water partition coefficient for the source zone (cm3water/g-soil)
V
= Volatilization rate constant for the source zone (day -1)
b
= Bulk density of the soil (g-soil/cm3-soil)
The volatilization rate constant, V, can be estimated from:
V =
2 DwT , S
4d 2
(11)
where:
DwT,S = Total effective diffusion coefficient defined in terms of a chemicals
concentration gradient in water (cm2/day)
DwT,S is defined mathematically as:
DwT , S = Dw, eff + K a Da , eff + K o Do, eff
(12)
w3.33
a 3.33
o 3 .33
;
D
=
D
;
and
D
=
D
a
,
eff
am
o
,
eff
om
2
2
2
(13)
where Dwm , Dam , and Dom are the molecular diffusion coefficients (cm2/sec) of
the given chemical in the water, air, and residual oil phases, respectively. The
following values were used for benzene: Dwm = 1.10E-05 cm2/sec [9], Dam =
0.093 cm2/sec [9], Dom = 3.4E-05 cm2/sec [10]. In general, the contribution of
the oil phase diffusion term to DwT,S was not significant.
The expression for the first order volatilization rate constant, V, was determined
from the solution for transient diffusion at long times from a slab of thickness, d,
with a pore water concentration equal to zero at the top surface and zero flux at
the bottom surface [11]. We have therefore assumed that there is no additional
resistance to mass transfer at the soil-air interface. The boundary layer mass
transfer resistance will be small relative to the diffusion resistance in the soil, so
it is reasonable to neglect this resistance. If an overburden layer exists, then its
thickness can be added to the parameter d in Equation 11. This assumes that the
overburden properties are similar to the source region. This would be a
conservative assumption with respect to source depletion, since the partitioning
165
Chapter 11
would be lower in the overburden layer than in the source region due to the
presence of residual hydrocarbon in the source versus the overburden layer.
Biodegradation of benzene in subsurface soils can be limited by the mass
transfer of oxygen. As a result, the first step in evaluating attenuation of
benzene due to degradation was to determine whether there would be sufficient
oxygen present from the bottom of the contaminated source region downward
through the unsaturated zone to the groundwater table. When oxygen
concentrations are above a threshold concentration, then a conservative first
order reaction rate constant can be used to estimate the attenuation of benzene in
the unsaturated zone. In developing the benzene RBSLs, a conservative first
order pore water-based biodegradation rate constant = 0.01 day-1 was assumed
when the source thickness was less than a critical value.1 For a sandy soil this
critical source thickness was estimated to be d 5 ft (calculations are not
presented here). A pore water-based biodegradation rate constant = 0.01 day-1
translates to a soil based degradation constant, unsat = 0.00094 day-1 for a sandy
soil with a volumetric water content, w = 0.094 cm3-w/cm3-soil.
Note that Equation (8) assumes that the DAFunsat is based on the maximum
aqueous concentration that will reach the groundwater table. It is therefore
conservative, since the average concentration in water observed over a typical
exposure period would be significantly lower.
Summary of DAFmix
The DAF in the mixing zone can be determined from the following equation:
DAF mix = 1 +
U gw gw
IL
(14)
where:
Ugw = Groundwater Darcy velocity (cm/year)
I = Infiltration rate of water through soil (cm/year)
L = Length of source area parallel to groundwater flow direction (cm)
166
Chapter 11
DAFsat
(Without Degradation)
1.0
1.1
2.2
4.2
11
41
163
167
Chapter 11
included (meaning DAFsat > 1), the RBSLs would be greater than those shown in
Figure 4.
100
Unweathered cond.
(median Ben conc.)
H=100 ft
H=50 ft
10
Unweathered crude
(median Ben conc.)
H=15 ft
H=5 ft
0.1
H=0
0.01
0.001
10
100
1000
10000
Figure 4. Comparison of soil RBSLs for benzene with a) predicted soil levels
for unweathered crude oils and condensates and b) actual soil levels after
emergency response activities at a few E&P sites. Open (white) symbols
represent detection levels for benzene. Soil RBSLs for benzene are for the
groundwater exposure pathway. Calculations were based on a sandy soil type
and include vadose zone attenuation due to volatilization and leaching from the
source and biodegradation beneath the source (see Figure 3). H = the distance
from the bottom of the contaminated soil zone to the groundwater table.
The calculation of RBSLs considers the following important factors: 1)
degradation in the unsaturated zone between the source of contamination and the
groundwater table, 2) source losses due to volatilization and leaching, and 3)
enhanced partitioning of benzene in soils due to the presence of TPH.
Parameters that determine the relative importance of these factors include the
depth to groundwater, the thickness of the source of contamination, and the soil
type. All of these three factors significantly contribute to higher acceptable
levels of benzene in soil than would be acceptable if attenuation and increased
partitioning in the vadose zone were not considered.
The RBSL curves presented in Figure 4 are based on a sandy soil type; a
biodegradation constant, unsat = 9.4E-04 day-1; a hydrocarbon impacted layer of
thickness, d=5 ft; and a groundwater standard = 0.010 mg/L. Values of other
parameters for determining the curves in Figure 4 are listed in Table 3.
168
Chapter 11
Unsaturated Zone
Layer Beneath Source
20
0.349(2)
0.094(3)
0.0
0.255
1.10E-05
0.093
2.20E-01
2.00E-03
83
1.73(2)
0.252
78
9.4E-04
-
169
Chapter 11
10,000 mg/kg results in an increase in the RBSL to 0.21 mg/kg. There is less
effect of TPH on RBSLs at greater depths, H, because higher TPH levels result
in relatively less attenuation from source losses.
Effect of Depth to Groundwater, H, on Benzene RBSLs-gw
When H = 0, there is no attenuation in the unsaturated zone (DAFunsat = 1). As H
increases, attenuation occurs between the bottom of the contaminated zone and
the groundwater table. This is a result of degradation over depth H and of losses
due to volatilization and leaching in the hydrocarbon impacted layer of
thickness, d. As H increases, attenuation increases, and the RBSL increases.
For d=5 ft, the increase in RBSL is due to increased degradation with
contributions from losses due to volatilization and leaching in the contaminated
region.
RBSLss
(mg/kg Soil)
484
290
174
170
Chapter 11
Summary
An improved approach to developing RBSLs for benzene in soil that are
protective of groundwater was developed that makes use of more realistic fate
and transport assumptions than are typically used in most Tier 1 calculations.
Attenuation effects were considered, including a conservative accounting of
minimal biodegradation in the unsaturated zone between the bottom of the
contaminated zone and the water table, as well as losses in the source of
contamination due to volatilization and leaching. In addition to these
attenuation considerations, the enhanced soil-water partitioning that occurs for
benzene due to the presence of TPH in soil was also considered. All of these
factors significantly contribute to higher acceptable levels of benzene in soil
than would be estimated if attenuation and increased partitioning in the vadose
zone were not considered. Additional attenuation in the saturated zone was not
considered but would further increase acceptable levels of benzene in some
cases.
Benzene RBSLs were found to depend on some key parameters: 1) depth to
groundwater (H), 2) thickness of oil impacted layer (d), and 3) level of TPH in
the soil. These parameters determine the amount of attenuation of benzene in
171
Chapter 11
the vadose zone and the decreased partitioning of benzene to soil when TPH is
present. An easy to use figure was presented that can be used to select the
appropriate benzene RBSL for an individual site if the above factors are known.
This figure illustrates that RBSLs increase by a factor of 10 to 1000 when
vadose zone attenuation and increased partitioning (water to soil) due to TPH
are taken into account.
The potential risk posed by benzene at E&P sites was also evaluated. Overall,
the major risk management concern for benzene at most E&P sites is likely to be
due to its potential to impact groundwater and not due to direct commercial
worker exposure to impacted surface soils. Benzene may present a risk to
groundwater at some E&P sites. The potential risk will depend on the type of
oil (crude oil or condensate) spilled, the depth to groundwater, the thickness of
contamination, the level of TPH in the soil, and the extent of weathering of
benzene from soils that results from any emergency response activities.
The benzene RBSLs presented in this chapter are illustrative of screening levels
that could be used in conjunction with TPH RBSLs to decide if further
corrective action is required at a given site. To use screening levels such as
these for benzene it is assumed that some response to a spill has already
occurred and that the extent of contamination has been delineated such that the
depth of contamination and level of contamination of benzene and TPH are
known.
References
1.
2.
3.
4.
5.
Rixey, W.G., Garg, S., and He, X., A Methodology for Accounting for the
Finite Leaching Characteristics of Contaminated Soils and Oily Wastes on
Groundwater Dilution and Attenuation, Environmental Engineering
Science, 17, 117-127 (2000).
172
Chapter 11
6.
nl, K., Kemblowski, M.W., Parker, J.C., Stevens, D., Chong, P.K., and
Kamil, I., A Screening Model for Effects of Land-Disposed Wastes on
Groundwater Quality, J. Contaminant Hydrology, 11, 27-49 (1992).
7.
van Genuchten, M.Th., and Alves, W.J., Analytical Solutions of OneDimensional Convective-Dispersive Solute Transport Equation, U.S.
Salinity Lab., Riverside, California, U.S. Department Agric.-Agric. Res.
Serv., Tech. Bull. No. 1661, 149 pp (1982).
8.
9.
173
Chapter 12
Introduction
Metals and semi-metals are commonly found in the environment (hereafter
metals and semi-metals are simply referred to as metals). They are present in
the rocks, soil, and organic matter that are the building blocks for the earth.
Some metals (such as chromium, selenium, and zinc) are essential to life and
must be supplemented as trace elements in the diets of humans and animals.
However, adverse health effects may be produced in people or other
environmental receptors when they are exposed to metals at certain
concentrations and under certain exposure conditions. For example, millions of
people inhabiting regions having iodine-deficient soils in eastern Africa are
susceptible to goiter, while Itai-Itai disease in China is attributed to people living
in areas where soils are contaminated by cadmium-containing wastes [1]. Some
metals , such as arsenic, cadmium, chromium, and nickel, have been shown to
produce cancer in people under some exposure conditions. Some metals have
been shown to produce adverse reproductive and/or other types of health effects
in people (e.g., lead, cadmium, mercury) and other animals (e.g., mercury,
selenium).
The chemical species of a metal is important in determining a metals toxicity.
For example, the inorganic form of arsenic is believed to be the carcinogenic
form, while the organic forms are not. Also, hexavalent chromium is
carcinogenic, while trivalent and elemental chromium are not. Therefore,
knowledge of the specific form of metal in an environmental sample (such as an
oil, waste, or soil sample) is important for accurate risk evaluation. Although it
is well known that different chemical forms of the same metal have different
toxicities, the analytical methodologies commonly employed do not readily
distinguish between different metal species. As a result, risk evaluations for
metals are usually based on the very conservative assumption that any metal
detected in a sample is in a form that may produce toxicity. This assumption
results in an overestimation of the potential risks posed by metals in
environmental samples.
Metals are natural components of crude oil. The chlorophyll molecule in
decomposing vegetative matter loses magnesium during crude oil formation, and
the magnesium is replaced by vanadium or nickel. The amount of vanadium and
174
Chapter 12
nickel in crude oils is well documented and their relative abundance can be used
to identify the source and age of crude oils [2]. However, there is little
published information on other metals in crude oils that may be of concern to
human health and the environment.
In contrast, the metals content of drilling muds has been investigated by the
United States Environmental Protection Agency (USEPA) [3]. Drilling muds
contain elevated levels of some specific metals such as barium. The American
Petroleum Institute (API) has also analyzed drilling muds and other exploration
and production (E&P) wastes (e.g., tank bottoms and produced sand) for metals
[4].
Through a joint industry research project carried out by the Petroleum
Environmental Research Forum, a more thorough understanding of the types
and concentrations of metals in crude oil was gained. The identification of
particular species of metals in crude oils was not attempted. Commonly
employed analytical methodologies were used, and it was then assumed that the
metals detected were in a form that may produce toxicity for risk evaluation
purposes, consistent with standard risk assessment practice. Therefore, as
discussed above, the risk evaluation for the metals detected in this study is
conservative. The technical information obtained in this study is needed by
regulators, risk assessors, and site managers to implement risk-based decisions
at sites that have been impacted by crude oil.
175
Chapter 12
present in soils from those that may have originated from a crude oil spill, or
from other sources.
The USEPA and many state agencies have developed human health risk-based
screening levels for metals. These screening levels are not regulatory limits, but
are commonly used as Tier 1 screening tools for evaluating risks, as described in
Chapter 1. The USEPA has developed Preliminary Remediation Goals (PRGs)
[8] for evaluating potential human health effects at Superfund sites. The PRGs
for metals are provided in Table 1. As an example of a state regulatory
program, the Texas Natural Resource Conservation Commission (TNRCC) has
developed Protective Concentration Levels (PCLs) under the Texas Risk
Reduction Program (TRRP) and the lowest residential PCLs for all exposure
pathways are also listed in Table 1 [9].
PRGs and PCLs represent chemical concentrations in environmental media (soil,
water, and air) that are considered protective of humans, including sensitive
groups, over a lifetime of exposure. They can be used to evaluate potential
health risks, trigger further site investigation, and serve as initial remediation
goals. Chemical concentrations above PRGs or PCLs do not automatically
mean that a site must be cleaned up to ensure public health, but they do suggest
that further evaluation of the potential risks posed by site contaminants may be
in order.
The PRGs presented in Table 1 are based on human exposure pathways
involving direct contact with contaminated soil and they focus on incidental
ingestion, dermal contact, and inhalation of vapors or soil particles containing
the chemicals of concern. They are derived by combining current USEPA
toxicity values with standard exposure factors and they correspond to fixed
levels of risk [i.e., either a one-in-one million (1 x 10-6) cancer risk or a noncarcinogenic hazard quotient of one], depending upon the risk levels selected by
the regulatory agency. The Texas PCLs are based on the same direct contact
exposure pathways, but they include additional indirect exposure pathways such
as potential leaching to groundwater and uptake into garden vegetables grown in
residential surface soil. They correspond to a target cancer risk level of one-inone-hundred thousand (1 x 10-5) and a non-carcinogenic hazard quotient of one.
For antimony, barium, beryllium, cadmium, chromium (total), lead, selenium,
and thallium, the lowest residential PCL is based on the metal leaching to
groundwater. PCLs for direct human contact with impacted soil are higher than
the levels presented.
176
Chapter 12
Ag (Silver)
As (Arsenic)
Ba (Barium)
Be (Beryllium)
Cd (Cadmium)
Co (Cobalt)
Cr (Chromium
(Total))
Cu (Copper)
Hg (Mercury)
Mo (Molybdenum)
Ni (Nickel)
Pb (Lead)
Sb (Antimony)
Se (Selenium)
Sn (Tin)
Tl (Thallium)
V (Vanadium)
Zn (Zinc)
Texas
Lowest
Residential
PCL
Residential Industrial (mg/kg)
390
10,000
0.48
0.39
2.7
24
5,400
100,0007 440
API
Guidance
Level
(mg/kg)
NA
41
180,000
Lowest
Ecological
SSL6 (mg/kg)
2
ORNL-P
10 ORNL-P
500 ORNL-P
Mean
U.S. Soil
Conc.
[10]
(mg/kg)
ND
5.2
440
150
37
4,700
210 3
2,200
810
100,0007
450 3
1.8
1.5
1,300
2,400
NA
26
NA
15003
4
3
20
0.4
CCME-A
CCME-A
ORNL-P
ORNL-E
0.63
ND
6.7
37
NA
10
NA
NA
500
NA
10
NA
NA
2,900
23
390
1,600
400
31
390
47,000
5.2
76,000
610
10,000
41,000
750
820
10,000
100,0007
130
550
2.1
49
160
3
5.4
2.3
35,000
1.7
750
17
37 4
210
300
NA
100 5
NA
NA
60
0.1
2
30
50
5
1
2
1
17
0.058
0.59
13
16
0.48
0.26
.89
ND
NA
500
550
23,000
14,000
100,0007
290
2,400
NA
1400
2
50
ORNL-E
ORNL-E
ORNL-P
ORNL-P
ORNL-P
ORNL-P
ORNL-P
CCME-A
ORNL-P
CCME-A
ORNL-P
ORNL-P
La29B
(mg/kg)
NA
10
20,000 1a
40,000 1b
100,0001c
NA
10
NA
500
58
48
177
Chapter 12
cleaned up, but they do suggest that the chemical(s) should be further evaluated
in an ecological risk assessment.
Ecological SSLs may be specific to a particular type of ecological receptor such
as plants, invertebrates, or microbial communities, and they may be specific to a
particular type of land use. The lowest ecological benchmarks available from
published sources are shown in Table 1. Sources for ecological soil benchmarks
include the Oak Ridge National Laboratory (ORNL) soil benchmarks for soil
and litter invertebrates (e.g., earthworms), heterotrophic soil microbial
communities, and terrestrial plants [11,12]; Canadian Council of Ministers of the
Environment (CCME) remediation criteria for agricultural land-use [13]; and the
Dutch National Institute of Public Health and the Environment (RIVM)
Ecotoxicological Intervention values [14]. Note that ecological SSLs have not
been developed for specific hydrocarbons in soil (e.g., benzene and
naphthalene); however, some have been developed for hydrocarbon fractions of
a Canadian crude oil [15].
In any risk evaluation for metals, it is critical to consider the background levels
of metals in soils. A thorough investigation of the concentrations of metals in
uncontaminated background soils in the United States has been reported by the
United States Geological Survey [10]. The mean values reported for metals in
soils in the United States from this publication are also provided in Table 1.
178
Chapter 12
6
4
1
2 1
1
2
1
3
Figure 1. Number of crude oil samples (26 total) analyzed by the geographic
region from which they originated.
For all metals except mercury and selenium, a 1-gram aliquot of each crude oil
was weighed into a glass digestion flask and sealed with a glass watch cover.
The samples were moved to hotplates and then digested with concentrated, highpurity sulfuric acid (H2SO4), nitric acid (HNO3), and hydrogen peroxide. A
separate digestion was performed for selenium, with 1-gram subsamples of each
crude oil and high-purity HNO3 as the only oxidizing acid. For mercury
determinations, a 0.1-gram subsample of each crude oil was weighed into a glass
digestion tube and oxidized with concentrated, high-purity HNO3 and H2SO4.
Once the crude oil digestions were completed, the samples were placed in
graduated cylinders, diluted to 10 mL with reagent water rinses of the digestion
tubes, and stored for analysis in 15 mL polyethylene bottles.
Crude oil Standard Reference Materials (SRMs) with trace metal concentrations
certified at environmental levels do not exist; therefore, additional crude oil
subsamples were weighed and then spiked with the elements of interest. Acid
digestion was selected for crude oil decomposition rather than ashing techniques
for two reasons: (1) the lower temperatures used during acid digestion were less
likely to cause the loss of volatile elements, and (2) the risk of sample
contamination was less.
Metal concentrations of the digested crude oil samples, spiked samples, and
blanks were determined by one of four methods: flame atomic absorption
spectrometry (FAAS), graphite furnace atomic absorption spectrometry
(GFAAS), inductively coupled plasma-mass spectrometry (ICP-MS), or coldvapor atomic absorption spectrometry (CVAAS). Concentrations of zinc (Zn)
were measured by FAAS using a Perkin-Elmer Model 4000 AAS. Cadmium
179
Chapter 12
(Cd), cobalt (Co), chromium (Cr), copper (Cu), molybdenum (Mo), nickel (Ni),
and vanadium (V) concentrations were determined by GFAAS with a PerkinElmer Model 4000 AAS utilizing a HGA-400 graphite furnace and AS-40
autosampler. A Perkin-Elmer Model 5100 AAS with HGA-600 graphite
furnace and AS-60 autosampler was used to measure concentrations of silver
(Ag), arsenic (As), beryllium (Be), and selenium (Se) by GFAAS.
Concentrations of barium (Ba), lead (Pb), antimony (Sb), tin (Sn), and thallium
(Tl) were measured by ICP-MS using a Perkin-Elmer ELAN 5000 spectrometer.
Crude oil mercury (Hg) concentrations were determined by CVAAS using a
Laboratory Data Control Model 1235 Mercury Monitor.
Africa #1
Africa #2
Africa #3
Africa #4
Asia
Central America
Indonesia #1
Indonesia #2
Indonesia #3
Middle East #1
Middle East #1*
Middle East #2
North America #1
North America #2
North America #3
North America #4
North America #5
North America #5*
North America #6
North America #7
North America #8
North America #9
North America #10
North America #11
North Sea
South America #1
South America #2
South America #3
API
Gravity
35.0
37.1
37.8
33.3
46.4
20.9
19.4
32.3
31.9
32.5
32.5
33.5
29.3
36.1
40.8
39.2
22.6
22.6
40.7
28.1
13.3
43.4
30.7
26.0
19.5
12.0
19.2
16.2
Ag
0.07
0.11
0.15
0.09
0.11
0.07
0.16
0.14
0.23
0.14
0.14
0.16
0.17
0.30
0.10
0.11
0.13
0.15
0.20
0.15
0.28
0.21
0.19
0.09
0.05
0.08
0.14
0.18
As
0.19
ND
ND
ND
ND
0.17
ND
0.57
0.09
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.19
ND
0.23
ND
0.19
ND
ND
ND
Ba
ND
ND
ND
0.054
0.032
0.002
0.189
0.041
0.036
ND
ND
ND
0.003
0.018
ND
0.368
0.087
0.090
0.011
0.036
0.206
ND
0.006
0.014
0.124
0.015
ND
0.095
Be
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Cd
0.006
0.003
0.014
0.026
0.016
0.005
0.020
0.008
0.010
0.006
0.007
0.016
0.005
0.015
0.007
0.006
0.006
0.007
0.005
0.003
0.013
0.003
0.011
0.006
0.005
0.010
0.016
0.009
Co
0.44
ND
ND
0.38
ND
0.11
0.69
0.38
0.30
ND
ND
ND
0.02
0.02
0.02
ND
1.13
1.08
ND
ND
0.81
ND
ND
1.33
0.67
0.27
0.16
0.22
Cr
0.035
0.021
0.369
0.124
0.067
0.211
1.43
0.667
0.869
0.074
0.079
0.016
0.248
0.111
0.033
0.022
0.864
0.874
ND
0.167
0.398
0.095
0.329
0.379
0.069
0.117
0.088
0.214
Cu
0.031
0.023
0.058
0.083
0.069
0.031
0.052
0.147
0.098
0.024
0.028
0.046
0.048
0.133
0.050
0.079
0.173
0.165
0.059
0.046
0.148
0.012
0.241
0.234
0.055
0.068
0.026
0.071
Hg
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1.56
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
180
Chapter 12
Table 2 (Continued). Amount of metals in crude oils (data are in mg/kg oil).
Africa #1
Africa #2
Africa #3
Africa #4
Asia
Central America
Indonesia #1
Indonesia #2
Indonesia #3
Middle East #1
Middle East #1*
Middle East #2
North America #1
North America #2
North America #3
North America #4
North America #5
North America #5*
North America #6
North America #7
North America #8
North America #9
North America #10
North America #11
North Sea
South America #1
South America #2
South America #3
API
Gravity
35.0
37.1
37.8
33.3
46.4
20.9
19.4
32.3
31.9
32.5
32.5
33.5
29.3
36.1
40.8
39.2
22.6
22.6
40.7
28.1
13.3
43.4
30.7
26.0
19.5
12.0
19.2
16.2
Mo
0.49
0.50
0.60
0.43
0.58
4.01
0.85
0.41
0.41
0.47
0.45
0.87
0.41
0.31
0.35
0.48
0.54
0.53
0.62
0.31
0.71
0.41
0.53
1.91
0.78
1.29
0.71
0.97
Ni
3.20
0.33
0.87
7.28
0.08
53.5
24.1
9.39
4.62
4.32
4.48
4.32
12.3
4.30
5.09
2.28
50.4
51.8
0.30
14.1
55.9
0.05
9.97
57.8
6.87
93.0
36.4
50.3
Pb
0.008
0.010
0.014
0.018
0.009
0.008
0.025
0.028
0.015
0.009
0.011
0.025
0.024
0.101
0.005
0.045
0.006
0.005
0.035
0.038
0.111
0.005
0.069
0.018
0.018
0.022
0.020
0.149
Sb
ND
0.001
ND
ND
ND
0.036
0.010
0.005
0.010
0.008
0.011
0.019
0.003
0.002
ND
0.005
0.013
0.012
0.012
0.004
0.010
0.001
0.009
0.017
0.001
0.055
0.028
0.022
Se
0.05
0.03
N.D
0.02
0.04
0.23
0.04
0.04
0.03
0.08
0.08
0.14
0.12
0.27
0.06
0.04
0.52
0.52
0.05
0.12
0.44
N.D
0.13
0.31
0.13
0.46
0.24
0.43
Sn
2.72
0.11
0.81
1.50
9.66
1.23
3.26
0.44
1.30
0.90
0.91
2.41
2.33
0.42
0.31
2.43
0.27
0.27
0.54
0.11
0.11
0.14
0.49
2.26
0.04
0.08
0.18
1.45
Tl
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.004
ND
ND
0.004
ND
ND
ND
0.002
ND
ND
ND
ND
ND
V
0.13
0.47
1.3
0.91
1.2
320
1.4
1.2
0.15
21
21
7.2
33
20
1.0
4.6
45
44
0.44
40
100
0.36
0.66
120
42
370
250
250
Zn
4.06
2.30
3.23
2.50
4.81
0.58
0.63
6.04
1.28
2.61
2.61
4.21
2.30
3.70
8.42
3.96
3.54
3.70
2.04
2.89
0.59
ND
10.9
3.39
1.10
ND
0.66
ND
ND = Not Detected
* = Duplicate Analysis
181
Chapter 12
Table 3. Statistical summary of metals content of 26 crude oils (the data are in
mg/kg oil).
Mean
Minimum
Maximum
Detection Frequency
(# per 26 Oils)
Method Detection Level
Suggested USEPA
Reporting Limit
Mean
Minimum
Maximum
Detection Frequency
(# per 26 Oils)
Method Detection Level
Suggested USEPA
Reporting Limit
Ag
0.15
0.05
0.30
26
As
0.06
ND
0.57
7
Ba
0.052
ND
0.368
19
Be
ND
ND
ND
0
Cd
0.010
0.003
0.026
26
Co
0.27
ND
1.3
16
Cr
0.270
ND
1.43
25
Cu
0.081
0.012
0.241
26
Hg
0.06
ND
1.56
1
0.010
0.1
0.080
0.5
0.001
1
0.005
0.1
0.002
0.2
0.010
0.2
0.005
0.5
0.010
0.5
0.010
0.6
Mo
0.77
0.30
4.0
26
Ni
19.69
0.05
93.0
26
Pb
0.032
0.005
0.149
26
Sb
0.011
ND
0.055
21
Se
0.16
ND
0.52
24
Sn
1.37
0.04
9.66
26
Tl
0.000
ND
0.004
3
V
62.75
0.13
370.0
26
Zn
2.92
ND
10.9
23
0.020
0.2
0.020
1
0.001
1
0.001
1
0.02
1
0.010
---
0.002
1
0.020
0.5
0.080
1
ND = Not Detected
The mean values for the oils were less than 1.5 mg/kg for all metals except
nickel, vanadium, and zinc, which had mean values of 19.7, 62.8, and
2.9 mg/kg, respectively. Beryllium was not detected in any of the crude oils.
Only one oil from the San Joaquin Valley in the State of California contained
mercury at a concentration above the detection limit. Mercury in this oil has
been previously reported, and its presence is believed to be due to the proximity
of the oil reservoir to mercury (quicksilver) deposits [16]. Thallium was
detected in only 3 of the oils, and arsenic was detected in only 7 of the 26 crude
oils.
182
Chapter 12
greater than its PRG, and two contained vanadium at concentrations greater than
its PCL. For the oil exceeding the arsenic PRG, the crude oil concentration in
soil that would result in an arsenic concentration exceeding the PRG would be
666,666 mg/kg, or 66.7% total petroleum hydrocarbons (TPH). Similarly, for
the oil containing the highest concentration of vanadium, the oil content in soil
would have to be more than 78.4% TPH in order to exceed the vanadium PCL in
Texas.
Therefore, it is highly unlikely that metals in soil from a single crude oil spill or
a single application of oil to the land surface would result in a significant risk to
human health. Sites that have received multiple applications of oil or oilcontaminated soils, such as a landfarm site, would require further study. Also,
E&P wastes may contain higher concentrations of some metals than their parent
crude oils due to corrosion processes, chemical additives, or metals in produced
water, and more data may be required to evaluate their risks.
183
Chapter 12
Lowest Ecological
Benchmark 1 (mg/kg)
5
ORNL-P
10
ORNL-P
500
ORNL-P
4
CCME-A
3
CCME-A
0.4
ORNL-E
20
ORNL-P
60
ORNL-E
50
ORNL-P
0.1
ORNL-E
2
ORNL-P
30
ORNL-P
1
ORNL-P
2
ORNL-P
1
ORNL-P
CCME-A
2
CCME-A
2
ORNL-P
50
ORNL-P
Maximum Concentration in
Crude Oil (mg/kg oil)
0.055
0.567
0.368
<0.005
0.026
1.43
1.33
0.241
0.149
1.56
4.01
93
0.52
0.296
0.004
9.66
370
10.9
184
Chapter 12
Conclusions
Twenty-six crude oils were analyzed for 18 metals. Evaluation of the human
health risk associated with soil containing crude oil showed that the potential
risk was not significant at total oil concentrations in soil well above current
management levels of 10,000 mg/kg TPH, used in many states. The amount of
metals in 10,000 mg/kg TPH would also be protective of soil invertebrates,
plants, and soil microbial communities. Vanadium may be of ecological concern
only in some heavy crude oils from Central and South America.
It is apparent that acceptable levels for crude oil in soil based on the potential
human health effects of metals are well above those that would be expected to
produce unacceptable aesthetic effects. Overall, the concentrations of these
metals in crude oils are unlikely to be a major risk management consideration at
crude oil spill sites. Therefore, routine analyses for metals in soils at crude oil
spill sites is not recommended.
References
1.
2.
3.
U.S. Environmental Protection Agency (USEPA), Technical Report Exploration, Development, and Production of Crude Oil and Natural Gas,
Field Sampling and Analysis Results, Office of Solid Waste and
Emergency Response, Publication 530-SW-87-005, Washington, D.C.
(1987).
4.
American Petroleum Institute (API), Oil and Gas Industry Exploration and
Production Wastes, API Document No. 471-01-09, Washington, D.C.
(1987).
5.
185
Chapter 12
6.
7.
8.
9.
10. Shacklette, H.T., and Boerngen, J.G., Element Concentrations in Soils and
Other Surficial Materials of the Conterminous United States, United States
Geological Survey Professional Paper 1270 (1984).
11. Efroymson, R.A., Will, M.E., and Suter, G.W., Toxicological Benchmarks
for Contaminants of Potential Concern for Effects on Soil and Litter
Invertebrates and Heterotrophic Process: 1997 Revision, ES/ER/TM126/R2, Oak Ridge National Laboratory, Oak Ridge, Tennessee (1997).
12. Efroymson, R.A., Will, M.E., Suter, G.W., and Wooten, A.C.,
Toxicological Benchmarks for Contaminants of Potential Concern for
Effects on Terrestrial Plants: 1997 Revision, ES/ER/TM-85/R3, Oak Ridge
National Laboratory, Oak Ridge, Tennessee (1997).
13. Canadian Council of Ministers of the Environment (CCME), Interim
Canadian Environmental Quality Criteria for Contaminated Sites, CCME
EPC-CS34, Winnipeg, Manitoba (1991).
14. van den Berg, R., Denneman, C.A.J., and Roels, J.M., Risk Assessment of
Contaminated Soil: Proposal for Adjusted, Toxicologically Based Dutch
Soil Clean-up Criteria, Arendt, F., Annokkee, G.J., Bosman, R., and van
den Brink, W.J. (eds), Contaminated Soils 93, Kluwer Academic
Publications, The Netherlands (1993).
15. Canadian Council of Ministers of the Environment (CCME), Canada-Wide
Standards for Petroleum Hydrocarbons (PHCs) in Soil: Scientific Rationale,
Supporting Technical Document, December 2000.
16. Bailey, W.H., Snavely, Jr., P.D., and White, D.E., Chemical Analyses for
Brines and Crude Oil, Cymric Field, Kern County, California, Short
Papers in the Geologic and Hydrologic Sciences , Articles 293-435
Geological Survey Research 1961, Paper 424-D, pp D-306-309.
186
Chapter 13
Application of Risk-Based
Decision-Making for International
Exploration and Production Site Management
Sara J. McMillen,1 Renae I. Magaw,1 Michele Emerson,2 and Lynn R. Spence3
1
Introduction
In general, the fundamental process of risk-based decision-making (RBDM) is
understood and accepted in the United States. In fact, as discussed in previous
chapters, several states have incorporated RBDM into their site remediation
programs. Howe ver, for most countries outside of North America and Europe,
the concept of RBDM is relatively new or unfamiliar. A few international
agencies, such as the Canadian Council of Ministers of the Environment [1] and
the Dutch Directorate-General for Environmental Protection, Department of Soil
Protection [2] have recently proposed or adopted risk-based approaches to
derive regulatory levels for total petroleum hydrocarbons (TPH). Appropriately,
they have utilized the same health risk-based concepts as those used in the
United States as they are equally applicable to sites anywhere in the world.
Moreover, in developing specific action levels for its program, each agency has
incorporated specific assumptions that reflect conditions in its own part of the
world to ensure that they are protecting the health of their citizens.
However, in other areas of the world this is not the rule. Most regulatory
programs appear to more commonly set cleanup criteria based on a review of
levels that have been set by North American or European agencies, rather than
on a site- or even country-specific basis. The assumptions that were used by the
North American or European agencies in setting cleanup levels are rarely
reviewed to determine whether they are indeed appropriate for the new
application in a given country. It is incorrect to merely assume that United
States or European calculations, which reflect potential exposures in those parts
of the world, are necessarily appropriate for other parts of the world. In order to
develop meaningful regulatory programs and cleanup levels, adequate
consideration must be given to factors such as differences in lifestyles, climate,
geology, etc., which apply to different parts of the world. This chapter presents
two examples in which RBDM and risk-based screening levels (RBSLs) were
applied to exploration and production (E&P) sites in Nigeria and Indonesia.
These examples demonstrate some of the important factors that must be
considered in applying RBDM at international locations.
187
Chapter 13
188
Chapter 13
Secondary
Source
Surface
Soils
Crude Oil
Terminal
Transport
Mechanism
Suspension
of Particulates
Subsurface
Soils
Volatilization
Groundwater
Infiltration,
Leaching
Surface
Water
Exposure
Media
Soil
Air
Outdoor
Groundwater
Surface
Water
Exposure
Pathway
Ingestion
Dermal contact
Inhalation
Inhalation
Receptor
On
sit
eW
ork
Of
er
fsit
eW
On
ork
er
sit
eR
es
id
Of
fsit ent
e
Re
sid
en
t
Primary
Source
X
X
X
Ingestion
Dermal Contact
Consumption
of Fish
189
Chapter 13
Parameter
Lifetime
Body Weight
Exposure Duration
Exposure Frequency
Soil Ingestion Rate
Inhalation Rate
Soil to Skin Adherence Factor
Skin Surface Area Exposed to Soil
Units
Years
kg
Years
d/Year
mg/d
m 3/d
mg/cm2
cm2
Resident
Adult
52
70
24
365
NA
22
NA
NA
Resident
Child
52
15
6
365
NA
10
NA
NA
Worker
52
70
25
183
100
22
0.2
7,000
NA = Not applicable, because inhalation of vapors and soil particles is the only
exposure pathway.
Table 2. Media and source geometry parameter values used for the RBSL
calculations.
Parameter
Lower Depth of Surficial Soil Zone
Fraction Organic Carbon in Soil
Particulate Emission Rate
Wind Speed Above Ground Surface in Outdoor
Air Mixing Zone
Width of Source Area in Major Direction of Wind
Outdoor Air Mixing Zone Height
Volumetric Air Content in Vadose Zone Soils
Total Soil Porosity
Volumetric Water Content in Vadose Zone Soils
Soil Bulk Density
Averaging Time for Vapor Flux
Units
cm
g OC/g Soil
g/cm 2-s
cm/s
Value
100
0.002
2.3E-12
225
cm
cm
(cm 3 Air)/(cm 3 Soil)
(cm 3 Voids)/(cm 3 Soil)
(cm 3 H2O)/(cm 3 Soil)
g/cm 3
s
1500
200
0.26
0.38
0.12
1.7
= Exposure Duration
(see Table 1)
Two receptors were considered for the generic Nigerian crude oil terminal
described above: an onsite worker and a resident adjacent to the site. The
worker is assumed to be at the site half of the year (approximately 183 days).
For many workers in the United States , the exposure time (used in the inhalation
calculations) is assumed to equal 8 or 9 hours per day (hr/day). Because the
onsite workers in this example are at the site 24 hr/day, the exposure time was
adjusted to 24 hr/day. It is possible that the workers could come in direct
contact with residually impacted soil, i.e., soil containing oil that has been left in
place after emergency response cleanup has occurred. The RBSLs developed in
this paper are for long-term worker exposures to oil impacted soil. The
remediation worker who handles the spill initially, but only over a short period
of time, is not addressed in this analysis.
190
Chapter 13
The exposure pathways considered for the workers are incidental ingestion of
soil, dermal contact with soil, and inhalation of soil vapor and particulate
emissions. While on-duty, the workers wear full protective clothing and it is not
likely that incidental ingestion and dermal contact would actually occur. These
exposure pathways were included to take into consideration the times when
workers may be off-duty and wearing lighter clothing, playing games like
soccer, or doing other miscellaneous activities around the site. The parameter
values assumed for the workers are shown in Table 1. The average lifespan in
Nigeria was obtained from the international database of the United States
Census Bureau [7]. The soil ingestion rate of 100 mg (per day) is twice the
value normally used in the United States for industrial scenarios, but it seems
appropriate given that the Nigerian workers are at the site for a much longer day
than workers in the United States . The skin surface area of 7,000 cm2 assumes
that the worker typically wears shorts and a short-sleeved shirt when off-duty,
but still on site [8]. The inhalation rate of 22 m3/day is recommended by the
World Health Organization [9] and is similar to the value typically used in the
United States . [8]. Worker exposure due to inhalation of indoor air is not
included in this example because of the many site-specific parameters that must
be investigated prior to calculating RBSLs for this pathway.
The resident is assumed to live at the sites property boundary. Because spills
occurring at the site are assumed to be contained within the property boundary,
the resident will not come in direct contact with the oil. Vapors and soil
particulates covered with oil may be carried by the wind to the residences, and
therefore the exposure pathways considered for the resident will be the
inhalation of vapor and dust emissions. When calculating risk from noncarcinogenic chemicals, the degree of risk is calculated by comparing the
estimated average daily dose with an acceptable dose (the reference dose, or
RfD). For residential exposures to non-carcinogenic chemicals, it is standard
risk assessment practice to assume that the receptor is a child. Children
typically have high inhalation and ingestion rates relative to their body weight,
and therefore they are usually the receptor with the greatest potential risk
because they can experience the highest average daily dose. The exposure
parameters used in this analysis are for a child 1 to 6 years in age, and they are
shown in Table 1.
Nigeria Crude Oil RBSLs
Five crude oils from Nigeria were analyzed using the modified TPHCWG
analytical method (see Chapter 4). Their American Petroleum Institute (API)
gravities ranged from 24 to 38. The calculated TPH RBSLs for noncarcinogenic effects are shown in Table 3 and they are in units of mg -TPH per
kg-soil (mg/kg). For onsite workers, the RBSLs for surface soil exposure range
from 38,000 to 45,000 mg/kg. These levels of crude oil will not result in
191
Chapter 13
significant risk to adult workers who have direct contact with oil impacted soil.
For offsite residents, the surface soil RBSLs range from 51,000 to 62,000.
These levels of crude oil will not result in significant risk to children from
inhalation of petroleum vapors or oily soil particles. (Note that if a spill were
not contained within the terminal boundaries so that residents might be exposed
to surface soils, the RBSLs could be lower by at least a factor of 10.)
For onsite workers and offsite residents the subsurface soil RBSLs are RES.
The term RES, for residual saturation, is used to indicate that the oil does not
pose a significant risk of adverse health effects even at residual levels in soil.
(The RES concept applies only for the leaching to groundwater and volatile
exposure pathways.)
Table 3. TPH RBSLs for Nigerian oils at a terminal site.
Oil Location/
Source
Crude Oil #1
Crude Oil #2
Crude Oil #3
Crude Oil #4
Crude Oil #5
Surface Soil
Offsite
Onsite Worker
Resident Soil
Soil RBSL
RBSL (mg/kg)
(mg/kg)
51,000
38,000
56,000
43,000
60,000
45,000
57,000
44,000
62,000
41,000
Subsurface Soil
Offsite
Onsite
Resident Soil
Worker Soil
RBSL (mg/kg) RBSL (mg/kg)
RE S
RES
RES
RES
RES
RES
RES
RES
RES
RES
RBSLs were not calculated for the potential fish ingestion pathway identified in
the CSM. The only chemicals of potential concern in Nigerian crude oils that
could be taken up by fish and transferred to people consuming the fish are the
polyaromatic hydrocarbons. However, these chemicals are relatively insoluble
in water and would not be expected to be dissolved in groundwater and
transported to the river as a result of an onsite spill.
192
Chapter 13
The natural ecosystem in Sumatra has been disturbed in some areas due to
logging, palm plantations, farming, and industrial activities. For these areas, it
is most appropriate to consider the risk to human health along with current and
future land use when setting cleanup levels or RBSLs. Therefore, RBSLs were
derived for specific oil types and considering specific land use (e.g., unrestricted
versus industrial or indemnified areas).
The science behind assessing risk to human health is well developed and has
been described at length in this book. However, data on the impact of
hydrocarbons in soil to specific ecological receptors (plants, animals, and
insects) within an ecosystem are very limited, and the data that are available are
largely from studies conducted with North American or European species.
Therefore, it is impossible at this time to set acceptable TPH limits for
protecting ecological receptors in undisturbed tropical rainforest ecosystems.
Site Description
Some oil production areas in Sumatra have been restricted to oilfield operations
only and are termed indemnified or restricted use. These areas can therefore
be evaluated much as industrial or commercial sites are evaluated in the United
States. However, oil production may also co-exist with other land uses such as
farming, residential areas, or plantations. These land uses were evaluated as
unrestricted or similar to United States residential land use scenarios.
Figure 2 illustrates a CSM for an unrestricted land use in Sumatra.
Primary
Source
Secondary
Source
Surface
Soils
Crude Oil
Production
Sites
Subsurface
Soils
Groundwater
Transport
Mechanism
Exposure
Media
Exposure
Pathway
Receptor
Onsite
Resident
Suspension
of Particulates
Volatilization
Infiltration,
Leaching
Surface
Water
Soil
Air
Outdoor
Groundwater
Surface
Water
Ingestion
Dermal contact
Inhalation
Inhalation
X
X
X
Ingestion
Dermal Contact
X
X
Ingestion
193
Chapter 13
Minas
500000
450000
400000
350000
300000
250000
200000
150000
100000
50000
>C
44
>6
-8
Ali
ph
atic
>8
s
-10
Ali
ph
ati
>1
cs
0-1
2A
liph
atic
>1
s
2-1
6A
liph
ati
>1
cs
6-4
4A
liph
ati
cs
>7
-8
Ar
om
atic
>8
s
-10
Ar
om
>1
ati
cs
0-1
2A
rom
atic
>1
s
2-1
6A
rom
atic
>1
6-2
s
1A
rom
atic
>2
s
1-4
4A
rom
atic
s
Figure 3. The TPH fractions [aliphatic and aromatic] and vacuum residuum
>C44 fraction for Minas and Duri oils.
RBSL Calculations for Crude Oils in Soil
The exposure assumptions commonly used in the United States were reviewed
for their suitability for deriving TPH RBSLs to be used in Indonesia. Very little
published information is available for developing Indonesia-specific exposure
parameters. A few exposure parameters were modified based on World Health
Organization information or site specific measurements s(uch as the fraction
organic carbon in soil). These are described in Table 4.
194
Chapter 13
Table 4.
RBSLs.
Parameter
Body Weight
Abbrev. Units
BW
kg
200
50
300
100
10
20
15
22
L/Day
1.5
Skin Surface
Area
SA
cm2/Day 2,900
3,160
3,900
4,100
Exposure
Frequency
EF
Days/
Year
350
250
365
365
g/g
0.01
0.01
0.002
0.002
mg/Day
United States
Sumatra Indonesia
Residential Comm./ Residential Comm./
(Child)
Indust. (Child)
Indust.
15
70
12
60
195
Chapter 13
supply. These RBSLs apply to TPH concentrations in surface soils, roads, oil
spills, or oily pit sites that are located on or near places where people live.
The RBSLs for unrestricted land use are low due to the living conditions in a
largely rural, developing region like Sumatra. The limiting pathways are dermal
exposure, soil ingestion, and soil particle inhalation. Leaching to groundwater
and volatile emissions will not usually be pathways of concern due to the low
concentratio ns of benzene (<50 mg/kg oil) and other low molecular weight
aromatic hydrocarbons in both Duri and Sumatra Light crude oils.
For restricted land use the RBSLs are much higher. The TPH RBSL for Duri
crude oil is 18,000 mg/kg TPH and is 30,000 mg/kg TPH for Sumatra Lighttype crude oils . This is due to lower amounts of exposure for industrial oil field
workers.
Conclusions
The RBDM process is a scientifically defensible, flexible, yet standard process
that can be used to develop international RBSLs. Simply adopting cleanup
standards that have been developed by European or North American regulatory
agencies for conditions in their countries will not provide meaningful standards
for developing countries.
At a generic terminal site in Nigeria, a RBDM approach is used to determine
potential receptors and complete exposure pathways. In the example presented,
TPH RBSLs are needed for three commercial worker pathways and two
residential exposure pathways. Similarly, example RBSLs were developed for
use in Sumatra Indonesia based on two land use scenarios. For both countries,
exposure parameters were modified to be protective of the way people live and
may be exposed to chemicals in their environments. In cases where site
conditions exceed these RBSLs, they can be used by a site manager as cleanup
levels or a more detailed Tier 2 or 3 risk assessment could be performed. It may
also be possible to reduce risk to human health by taking actions to eliminate
some of the pathways of concern.
References
1.
2.
196
Chapter 13
3.
4.
5.
Vorhees, D., Gustafson, J., and Weisman, W., Human Health Risk-Based
Evaluation of Petroleum Contaminated Sites: Implementation of the
Working Group Approach, TPHCWG, Vol. 5, Amherst Scientific
Publishers, Amherst, Massachusetts (1999).
6.
McMillen, S.J., Magaw, R.I., Kerr, J.M., Sweeney, R.E., Nakles, D.V., and
Geiger, S.C., A New Risk-Based Approach to Establish Cleanup Levels
for Total Petroleum Hydrocarbons, presented at the International
Petroleum Environmental Conference, Houston, Texas, November 1999.
7.
United States Census Bureau (USCB), International Data Base, Year 2000
Average Life Expectancy for Both Sexes in Nigeria (2000).
8.
9.
197
Appendix
Number
1.2
1.3
1.4
198
Appendix
1.6
1.7
199
Appendix
1.8
1.9
200
Appendix
No
GC Fingerprint
Similar for all
Samples?
No
2.0
Report
Results
Fractionate All
Samples
TPH Below
Action Level
Criteria?
Yes
Yes
Fractionate
Representative
Sample
Quantitate
TPH and BP
Distribution
Quantitate
Aliphatic and
Aromatic
Fractions
GC/FID of methylene
chloride (Alumina) or 1:1
methylene
chloride:Acetone (Silica)
eluate for Aromatics
Summary of Method
2.1
201
Appendix
2.3
2.4
202
Appendix
3.0
Definitions
3.1
3.2
3.3
3.4
203
Appendix
than the n-alkanes of the same carbon number. Thus, there are
some C8 hydrocarbons that elute before n-heptane and there
are some that elute after n-octane, including the aromatics
ethylbenzene and the xylenes.
This method allows for data reporting between each carbon
range or for reporting within wider carbon ranges depending
on data quality objectives. The TPHCWG has defined
fractions based on different properties that affect the fate and
transport and/or toxicity of petroleum hydrocarbon
components.
4.0
3.5
3.6
3.7
3.8
Interferences
4.1
204
Appendix
5.0
4.2
4.3
4.4
The toxicity of the reagents used in this method has not been
precisely defined. However, each chemical compound should
be treated as a potential health hazard. Exposure to these
chemicals should be reduced to the lowest possible level by
whatever means available. The laboratory is responsible for
maintaining a current awareness file of Occupational Safety
and Health Administration (OSHA) regulations regarding the
safe handling of the chemicals specified in this method. A
reference file or Material Safety Data Sheets (MSDS) should
also be made available to all personnel involved in the
chemical analysis. Additional references to laboratory safety
should be available and should be identified for use by the
analyst.
205
Appendix
6.0
Apparatus
6.1
Glassware
6.1.1
6.1.2
6.1.3
Vials
6.1.3.1 10 to 40 mL glass vials with Teflon-lined
screw caps.
6.1.3.2 2 mL GC autosampler vials with Teflonlined crimp caps.
6.1.4
6.1.5
6.1.6
6.2
Microsyringes: 10 L to 1000 L.
6.3
6.4
Vortex mixer.
6.5
6.6
Drying oven.
6.7
Gas Chromatography
6.7.1
206
Appendix
Recommended Columns
6.7.2.1 25/30 m x 0.25 to 0.53 mm ID fused silica
capillary column with 0.25 to 1.5 m film
thickness (methyl silicone) or equivalent.
Low bleed columns are preferred. Examples
include MS-007 (Quadrex), DB-1 (J&W),
and RTX-1 (Restek). DB-5 (J&W) may be
used.
6.7.2.2 Other columns may be used if the elution of
the compounds is based on boiling point.
Capillary columns are recommended. See
Section 9.3.2 for GC performance criteria.
7.0
6.7.3
6.7.4
Reagent grade,
7.2
7.3
7.4
Purify by
207
Appendix
7.4.2
208
Appendix
8.0
8.2
209
Appendix
9.0
~ Boiling
Point, C
69
98
126
151
174
196
216
236
253
270
287
302
316
329
343
n-Alkane
Marker
n-C21
n-C22
n-C23
n-C24
n-C25
n-C26
n-C27
n-C28
n-C29
n-C30
n-C31
n-C32
n-C33
n-C34
n-C35
~ Boiling
Point, C
357
369
380
391
402
412
422
431
441
450
458
467
474
481
499
Procedures
9.1
Sample Extraction
9.1.1
210
Appendix
211
Appendix
9.1.2
212
Appendix
9.2.2
213
Appendix
9.2.3
9.2.4
9.2.5
214
Appendix
9.3
9.2.6
9.2.7
Gas Chromatography
9.3.1
215
Appendix
to
216
Appendix
Before establishing windows for integration using nalkanes as markers, make sure that the GC system is
within optimum operating conditions. Make three
injections of the boiling point distribution standard
defined in 7.4.2 throughout the course of a 72-hour
period.
9.4.2
9.4.3
9.4.4
9.4.5
9.4.6
217
Appendix
9.5
9.4.7
9.4.8
218
Appendix
9.5.2
Aromatics
>C6 to C7 (Benzene)
>C7 to C8 (Toluene)
>C8 to C10
>C10 to C12
>C12 to C16
>C16 to C21
>C21 to C35
219
Appendix
StandardDeviation of 5 CFs
x 100
Mean of 5 CFs
220
Appendix
R1 - R2
x 100
Ravg
where:
R1 = Standard value or average CF
R2 = Calculated value or CF
Ravg= (R1 + R2)/2
9.5.2.5 Calibration of Selected Target Analytes:
Selected components (target compounds
such as benzene, toluene, ethylbenzene,
xylenes, PNAs, n-alkanes) can be measured
individually if desired.
Assuming an
221
Appendix
9.6.2
222
Appendix
9.6.3
9.6.4
9.6.5
223
Appendix
9.7
9.8
9.7.2
9.7.3
9.7.4
9.7.5
Calculations
9.8.1
Cc x Vt x D
1 mg
x
Ws
1000 g
224
Appendix
where:
Cs = Concentration of TPH, hydrocarbon range or specific
analytes (mg/kg or mg/L).
Cc = Concentration from calibration curve in g/mL. (If CF is
used for calculations, this value is area calibration/CF).
Vt = Volume of extract (mL).
D = Dilution factor, if dilution was performed on the sample
prior to analysis. If no dilution was made, D = 1,
dimensionless.
Ws= Weight of sample extracted (kg). If a water sample, then
the units are L.
9.8.2
9.9
10.0
Quality Control
10.1
225
Appendix
10.1.2
10.1.3
10.1.4
10.1.5
If any of the criteria in 10.3 and 10.4 are not met, the
problem should be corrected before samples are
analyzed.
10.1.6
10.1.7
226
Appendix
10.2
10.3
Minimum Instrument QC
10.2.1
10.2.2
10.2.3
10.2.4
10.2.5
227
Appendix
10.3.3
228
Appendix
229
Appendix
230
Appendix
11.0
11.1.2
231
Appendix
232
Appendix
12.0
13.0
Method Performance
12.1
The method has been applied to the analysis of neat crude oil,
gasoline, JP-4, and diesel. In addition, the method has been
used for the analysis of soil samples impacted with crude oil
and with petroleum products with different degrees of
weathering. Recoveries are typically 80% or better for most
samples.
12.2
12.3
Pollution Prevention
13.1
13.2
233
Appendix
14.0
Waste Management
14.1
15.0
References
1.
USEPA, Test Method for the Evaluation of Solid Waste (SW846), 3rd Edition (1996).
2.
3.
4.
5.
6.
7.
USEPA Test Method for the Evaluation of Solid Waste (SW846), Method 3611B, Alumina Column Cleanup and Separation
of Petroleum Wastes (1996).
8.
USEPA Test Method for the Evaluation of Solid Waste (SW846), Method 3630C, Silica Gel Cleanup (1996).
9.
234
Appendix
235
Glossary of Terms
Glossary of Terms
Acute: Rapid. An acute exposure is of short duration.
Aliphatic: Organic compounds which are non-aromatic and non-polar, such as
alkanes and alkenes.
Aliquot: A precise portion.
Alkane: A saturated hydrocarbon, containing only hydrogen and carbon in
single bonds. Synonymous with saturate fraction, paraffins, and naphthenes.
Alkene: Hydrocarbons containing carbon-carbon double bonds. Alkenes are
refinery by-products of the cracking of oils.
Aromatic: Unsaturated (double-bond) cyclic hydrocarbons containing the
benzene ring as the basic structural unit of the molecule.
API Gravity: The American Petroleum Institute method for specifying the
specific gravity of crude oil.
Benzene Ring: The simplest aromatic compound, containing one, 6-membered
carbon ring with double and single carbon-carbon bonds.
Biochemical: Pertaining to the chemistry of biological processes.
Biophysical: Pertaining to the physical nature of biological processes.
Cancer Slope Factor: A conservative estimate of the incremental probability of
a person contracting cancer from a unit intake of a chemical over a lifetime.
Carcinogen: A chemical, biologic, or physical agent that can cause cancer.
Chemical of Potential Concern: A chemical that has the potential to negatively
impact human health and the environment at a site.
Chronic: Occurring over a long period. Chronic exposures generally occur over
a long period of time. USEPA defines chronic exposures as those that may last
at least 10% of an average lifetime (i.e., 7 years or more).
Comprehensive Environmental Response and Contingency Liability Act
(CERCLA): 1980 United States federal law authorizing identification and
remediation of unsupervised hazardous waste sites and spill reporting (also
called Superfund).
Detection Limit: The minimum concentration of a chemical that can be
detected using a specific measurement procedure and laboratory equipment.
Dose-Response: The measurement of the response of an organism to a dose of
chemical.
Eluent: The mobile liquid phase used to separate chemical fractions of oil
during liquid chromatography.
Exposure: The means by which receptors come in contact with chemicals.
236
Glossary of Terms
237
Glossary of Terms
238
Glossary of Terms
239